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

J. Biol. Chem., Vol. 276, Issue 35, 33129-33138, August 31, 2001

This Article Abstract Full Text (PDF) All Versions of this Article: 276/35/33129    most recent M105400200v1 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 Wong, P. M. C. Articles by Chung, S.-W. Search for Related Content PubMed PubMed Citation Articles by Wong, P. M. C. Articles by Chung, S.-W. Social Bookmarking                      What's this?

A Single Point Mutation at the 3'-Untranslated Region of Ran mRNA Leads to Profound Changes in Lipopolysaccharide Endotoxin-mediated Responses^*

Peter M. C. Wong, Quan Yuan, Hong Chen, Barnet M. Sultzer, and Siu-Wah Chung

From the Department of Pathology & Laboratory Medicine, Fels Institute, Temple University School of Medicine, Philadelphia, Pennsylvania 19140

Received for publication, June 12, 2001

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

By functional cDNA expression cloning, we have previously established that Ran is important in lipopolysaccharide (LPS) signaling. This was achieved by functional comparison between two cDNAs, differing by a single base substitution within the 3'-untranslated region of the cDNA. This point mutation results in a striking RNA conformational change. No dramatic difference in total RNA at steady state could be found between the two molecules. However, at the protein level, RanC/d (from 870C mRNA) was 5-10-fold higher than RanT/n (from 870T mRNA) and this difference was not observed in non-hematopoietic cells transduced with the same vectors. This tissue-specific difference correlated with a difference in LPS endotoxin responses in corresponding hematopoietic cells. Importantly, the amounts of Ran- C/d and RanT/n proteins were similar initially but the difference became obvious with time. Both Ran proteins migrated from the cytoplasm to the nucleus, but Ran from RanC/d migrated faster than that of RanT/n. RanT/n protein preferentially remained in the cytoplasm and its overall amount was reduced at steady state, consistent with its degradation by intracellular proteases known to be involved in LPS-mediated signal transduction. As the two proteins are identical, the faster RanC/d nuclear localization and a preferred initial cytoplasmic RanT/n distribution suggest a difference in mRNA intracellular localization between the two molecules, as dictated by their RNA structural difference. By pulse-chase experiments, RanC/d proteins are more resistant to degradation than RanT/n protein; there also appear to have two populations of RanT/n proteins, one may reside in the cytoplasm and the other, in the nucleus. More RanC/d GTPase accumulated in the nuclei would conceivably alter the potency of signal transduction and therefore down-modulate LPS-mediated biological responses.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Endotoxin lipopolysaccharide (LPS)¹ is a major component of Gram-negative bacterial membrane, which accounts for more than 50% of all cases of septic shocks in humans in certain populations. The genetic basis for a variety of physiological and pharmacological host responses against bacterial endotoxin was recognized in the late sixties, when Sultzer (1) observed that C3H/HeJ inbred mice have a natural resistance to endotoxin challenge as compared with other mice. Subsequent breeding analyses by crossing C3H/HeJ mice with other sensitive stains suggested that the B cell mitogenic response to LPS may be governed by a single genetic locus composed of co-dominant alleles (2-6). Watson and associates (7) showed that this locus, Lpsⁿ, was on chromosome 4 and linked to the major urinary protein locus, Mup-1, but downstream from Mup-1 and the Lyb2;4;6 genes, which control B cell activation. Golde and Rosenstreich (43) surveyed 11 closely related C3H/HeJ mouse strains using the B cell proliferation assay and found that there were three levels of LPS responsiveness. They interpreted the result to mean that there were two distinct mutations. Earlier study of the pattern of inheritance to endotoxic shock by Sultzer (17) also suggested the presence of multiple genes in the control of sensitivity to endotoxic shock. By position cloning, the Tlr 4 gene has been localized to the Lps genetic locus defined by these cross-breeding analyses, and a missense mutation was identified in the third exon of the gene in the C3H/HeJ genome (8, 9). Toll-like receptors have been shown to be important in innate immunity not only initially in Drosophila but also in mammals (10-22). The involvement of as many as nine or more Toll-like surface receptors in the mammalian system therefore suggests that multiple signal transduction pathways are operative in innate immune responses, which characteristically result in the production of proinflammatory cytokines, further implying that these pathways may merge at some point inside the cytoplasm or the nucleus.

Using a functional cDNA expression cloning strategy, Kang et al. (23) isolated a cDNA whose expression in C3H/HeJ splenic B cells confers the ability to respond to LPS. The assay employed in that study was one of the functional B cell mitogenic assays originally used to define the defect in C3H/HeJ mice and the genetic basis for LPS responses (1, 2). This gene, Ran, encodes for a small G protein, Ran GTPase, which has been shown to be involved in diverse biological functions. These include nuclear transport of macromolecular RNA and proteins, and most recently, in binding to microtubule spindle formation during mitosis, which are biological events consistent with LPS-mediated B cell mitogenesis (24-28). Wong et al. (29) have also characterized another cDNA, with a single base substitution, at position 870, in the 3'-UTR of Ran mRNA, RanC/d (C for cytidine, d for "deficient" biological phenotype), which affected LPS response significantly. Expression of the RanT/n (T for thymidine, n for "normal" biological phenotype) but not the RanC/d cDNA, after retroviral gene transfer into C3H/HeJ splenic B cells, conferred a proliferative B cells responsiveness to endotoxin LPS stimulation (23, 29). In addition to using the B cell mitogenic assay, the investigators also employed lethality as an assay for host response to endotoxin challenge to compare the biological consequences in endotoxin-sensitive mice after adenoviral gene transfer. While RanC/d could confer resistance to endotoxin challenge in sensitive mice, RanT/n could not. Subsequent analysis suggests that RanC/d could confer resistance of endotoxin challenge via down-modulation of TNF production by responsive macrophages in a dominant negative fashion upon LPS stimulation (30).

To further understand the mechanism underlying the biological differences between RanT/n and RanC/d, we undertook an additional series of experiments. In this study, we report significant differences in secondary and tertiary RNA structure between the two mRNAs. There was no major difference in RNA stability between the two mRNA species. RanT/n mRNA produces the same amount of protein initially as the RanC/d mRNA, but with time the overall level of RanT/n protein was reduced at steady state compared with RanC/d protein, which is a function of their differential intracellular localization. These differences correlated with the ability of the transduced macrophages to production proinflammatory cytokine TNF in their response to endotoxin LPS stimulation.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Animals, Cell Lines, and Virus Infections-- All mice were purchased from Jackson Laboratory (Bar Harbor, ME). All cell lines were described in our previous publications (31-33), and retrovirus infection and constructions were performed as described before.

HA-tagging of RanC/d and RanT/n cDNAs-- Two oligonucleotides (5'-TGAGTATCCGTACGACGTCCCAGATTATGCATCACTAGG-3' and 5'-TCACCTAGTGATGCATAATCTGGGACGTCGTACGGATAC-3') were synthesized (Life Technologies, Inc., Grand Island, NY) and annealed. These double-stranded oligonucleotides encode an epitope sequence (YPYDVPDYASL) derived from hemagglutinin (HA) of influenza virus. They also contain the cohesive ends for a restriction enzyme, Bsu36I, which allows for the ligation with plasmids pCD-RanC/d and pCD-RanT/n (23, 29) that were linearized with Bsu36I. As a result of the ligation, the HA epitope sequences and two extra amino acid residues, GE, were inserted into the Lps/Ran (or Ran) cDNA between positions Glu¹⁸⁶ and Val¹⁸⁷ of the Ran protein. The newly formed plasmids were named pCD-RanC/d-HA and pCD-RanT/n-HA. The accuracy of such insertion was confirmed by DNA sequencing (36).

RNase Protection Assay-- Total RNA was extracted from 1 × 10⁸ cells according to Chomczynski and Sacchí (37). Briefly, cells were lyzed in 10 ml of 4 M GTC solution (4 M guanidine thiocyanate, 25 mM sodium citrate, 2.5% sodium lauryl sarcosine, and 0.1 M 2-mercaptoethanol). After DNA shearing, the following reagents were sequentially added to the GTC lysate and thoroughly mixed between each step: 1 ml of 2 M sodium acetate, pH 4.0, 10 ml of water-saturated phenol, and 2 ml of chloroform. After an incubation on ice for 15 min, the tubes were centrifuged at 10,000 × g for 20 min. The top aqueous phase containing RNA was transferred into new tubes, mixed with equal volume of isopropyl alcohol, repelleted, and the pellet dissolved in 0.4 ml of 4 M GTC solution until use.

To generate a suitable Ran-HA RNA probe for RNase protection, we synthesized two primer oligonucleotides: 5'-TTGTTGCCATGCCTGCTCTTG-3' and 5'-GGTCATCATCCTCATCTGGGA-3'. They were used for PCR to amplify a 158-bp fragment targeting at the Lps-Ran sequences, position 527 to position 646 of the original Lps/Ran (or Ran) cDNA (29), giving a size of 119 bp. Inclusion of the 39-bp HA-epitope sequence produces a fragment of 158 bp (Fig. 3B). After PCR amplification, the 158-bp fragment was then cloned into pCRII (Invitrogen, Carlsbad, CA), a vector ideal for PCR cloning because of the AT overhangs. The insert was cloned into the plasmid in its antisense orientation with respect to transcriptional direction of the T7 promoter. Successful cloning was confirmed by DNA sequencing. The full size of the Ran-HA probe, including vector sequences, was 286 nt (Fig. 3B). The size of the protected fragment for Ran-HA was 158 nt, whereas the size of the protected fragments of the endogenous Ran was 87 and 32 nt. We also measured the level of actin RNA in each sample, using a plasmid pTri--actin-mouse purchased from Ambion (Austin, TX). In vitro transcription of the -actin mRNA was driven by the Sp6 promoter. The full size of the actin probe was 334 nt, the size of the protected fragment was 245 nt. The in vitro transcribed Ran-HA and -actin mRNAs were synthesized and DIG-labeled using the Sp6/T7 DIG RNA labeling kit purchased from Roche Molecular Biochemicals (Indianapolis, IN).

For RNase protection assay, we modified the procedure from two sources: one source derived from the Direct Protect^TM kit (Ambion), the other from Garica-Sanz and Mullner (38). Briefly, 50 µl of GTC RNA solution were mixed with 5-µl probe mixtures (0.4 ng/µl of DIG-labeled Ran-HA RNA or 0.1 ng/µl DIG-labeled actin RNA) and incubated at 37 °C overnight. After incubation at 37 °C, 0.5 ml of RNase T1 solution (400-800 units of RNase T1 from Roche Molecular Biochemicals, 10 mM Tris·Cl, pH 8.0, 0.3 M NaCl, and 5 mM EDTA) was added to each tube and incubated at 37 °C for 30 min. Twenty µl of 10% Sarkosol was added to each tube and mixed; then 10 µl of 20 mg/ml Proteinase K (Ambion) was added to each tube, throughly mixed, and incubated at 37 °C for 30 min. An equal volume of acid phenol/chloroform was used to extract the treated RNA solution. The top aqueous phase was transferred into new tubes and precipitated with an equal volume of isopropyl alcohol. The pellet was dissolved in 10 µl of denaturing gel loading buffer II (95% deionized formamide, 18 mM EDTA, 0.025% SDS, 0.05% xylene, and 0.05% bromphenol blue). The final RNA solution was heated at 90-95 °C for 3 min and cooled on ice. Three µl of the RNA solution were loaded into each lane of a 6% denaturing gel with 8 × 10-cm configuration (6% acrylamide with 5% cross-linking, 7 M urea, and 0.5 × TBE), which had been pre-electrophoresed at 230 volts for 30 min with 0.5 × TBE as the upper chamber running buffer and 2.5 × TBE as the lower chamber running buffer. Ten picograms of DIG-labled RNA molecular weight marker mixture, which was in vitro transcribed with incorporation of digoxigenin-UTP from the DNA template mixture (Ambion) according to the manufacutrers instructions, were loaded into lanes on each side of the gel. The gel was electrophoresed at 230 volts for about 1 h or until the bromphenol blue dye ran about 1 cm away from the bottom. The gel was lifted with a sheet of Whatman 3MM paper attached onto it and assembled onto a sandwich for Western blot transfer set up. The gel was transferred to a neutral nylon membrane in 0.5 × TBE as transfer buffer at 66-76 volts for 20-25 min. The membrane was air dried for a few minutes and UV cross-linked at the autocross-linking setting in Stratalinker (Stratagene). The membrane was blotted with anti-DIG antibody for detection of any DIG-labeled products, according to instructions in Ambion's kit.

To determine the half-life of Ran-HA mRNA in 70Z/3 cells, retroviral transduced 70Z/3 cells (N2-RanC/d-HA and N2-RanT/n-HA) at 1 × 10⁶/ml were stimulated with or without 1 µg/ml LPS for overnight. 5,6-Dichlorobenzimidazole riboside (Sigma) was added to the cultures to a final concentration of 100 µM. A 50-ml aliquot of cell culture was removed from culture flask at times 0, 2, 4, and 8 h, respectively. The cells were washed with PBS and lysed in 4 M GTC solution. The RNase protection assays were performed on these GTC lysates as described above. Once the x-ray film with proper density of each protected fragments was obtained, the film was scanned and analyzed with IP Lab Gel, a Gel Analysis and Densitometry system (Scanalytics, Inc. Fairfax, VA). In detail, the total pixels in each fragment was divided by the total pixels at time 0, giving rise to percentage of mRNA remaining at a particular time point. Three independent experiments were performed. The mean percentages of mRNA remaining from each experiment at different times were used for regression analysis in SigmaPlot program.

Western Blot Analysis and Immunofluorescence Studies-- Western blot analysis was carried out as described previously (39).

For immunofluorescence studies, primary peritoneal macrophages from C3H/HeOuJ and C3H/HeJ mice were primed with 3% thioglycollate medium for 3 days. They were collected and plated onto 8-chamber slides (Lab-Tek) at 5 × 10⁴/0.5 ml/well and incubated at 37%, 5% CO₂ for 6 h. The slides were washed with PBS to remove non-adherent cells and were infected with Ad5-RanT/n or Ad5-RanC/d viruses, each at a 1:10,000 multiplicity of infection (10⁹ plaque forming units/well) in 100 µl of R10 (RPMI 1640 supplemented with 10% fetal calf serum) for 6 h. To allow for expression of the transgene, another 400 µl of R10 were added to each well and incubated further for an additional 14 h. At this point, the culture was stimulated with or without 1 µg/ml LPS (Salmonella typhimurium), unless otherwise stated. The cells were then analyzed after another 2, 12, 24, or 48 h. The slides were then washed with PBS three times and dried for 30-60 min at room temperature. The cells were then fixed with 4% paraformaldehyde for 30 min and washed with PBS three more times. Blocking solution (5% goat serum, 1% bovine serum albumin, 0.05% Nonidet P-40, and 0.01% NaN₃ in PBS) was applied and the slides were incubated for 30-60 min at room temperature. We then added 200 µl per slide of rat anti-HA antibody (Roche Molecular Biochemicals, fresh preparation of 100 ng/ml in blocking solution) and incubated them for 2 h at 37 °C. The slides were washed with 0.05% Nonidet P-40/PBS once, PBS three times, for 4 min each time. Next, we added 200 µl of goat anti-rat IgG conjugated with biotin (Jackson ImmunoResearch Labs; 1:500 dilution in 2% goat serum PBS) to each slide and incubated them for 60 min at 37 °C. After 3 washes with PBS, streptoavidin-Cy3 (Jackson ImmunoResearch Labs; 1:2000 dilution in 1% bovine serum albumin/PBS) was added and they were incubated for 20 min at room temperature in the dark. After another 3 washes with PBS, Vectashield mounting medium was added and a coverslip was placed on top for immunofluorescence analysis using a fluorescence microscope (Nikon E800) equipped with Texas Red and 4,6-diamidino-2-phenylindole single pass filters.

Protein Degradation by Pulse-Chase Analysis-- Primary peritoneal macrophages were collected from thioglycollate-primed C3H/HeOuJ mice seeded into 6-well plates at 2.5 × 10⁶ per well and incubated at 37 °C for 3 h. Adherent cells were infected with viral supernatant at a 10⁴ multiplicity of infection. Twenty hours after infection, the cells were washed twice with PBS, incubated for 1 h at 37 °C in depletion media (methionine-free RPMI 1640 medium, and 2% dialyzed fetal calf serum (GIBCO)), and then pulse-labeled with 50 µCi of [³⁵S]methionine per nell 1 h at 37 °C. After labeling, the cells were washed twice with the complete R10 medium and then incubated with R10 medium for variable times (12).

These cells were then subjected to immunoprecipitation and fluorography. After washing twice, cells were then lysed in RIPA buffer (50 mM Tris·Cl, pH 8.0, 150 mM NaCl, 25 mM EDTA, pH 8.0, 1% Nonidet P-40, 0.5% sodium deoxycholate, 0.1% SDS) containing protease inhibitor mixture (Roche Molecular Biochemicals). Cell lysates were stored frozen at 70 °C until use. Samples for immunoprecipitation were normalized on total protein concentration measured by DC protein assay (Bio-Rad). 200 µg of total protein were incubated with anti-HA antibody (2 µg/ml, Roche Molecular Biochemicals) at 4 °C overnight. Then 10 µl of Protein G-agarose (Life Technologies, Inc.) was added and incubated at 4 °C for another hour. Immunoprecipitated pellets were washed four times with RIPA buffer then boiled in 2 × SDS sample buffer for 5 min. The supernatant was then subjected to SDS-polyacrylamide gel electrophoresis on 12% gels. For fluorography, the electrophoresed gels were fixed in the solution containing 25% 2-propanol and 10% glacial acetic acid for 1 h and treated with Fluoro-Hance (RPI) for 30 min. Gels were dried onto Whatman filter papers and exposed to FUJI image plate. The plate was scanned with FUJI BAS2000 and analyzed by FUJI BAS 2000 EWS software.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

RNA Structural Differences between RanT/n and RanC/d-- The two Ran mRNAs differs from each other only by a single point mutation at position 870 at the 3'-UTR (29). To examine if this difference transcribes to differences in RNA structure between the two molecules, we first compared the secondary structure of the two sequences using the Squiggles program of the Wisconsin GCG Sequence Analysis Package. As shown in Fig. 1, the structure of RanT/n mRNA is strikingly different from that of RanC/d mRNA. This difference is conspicuous in the region close to the point mutation and is absent in regions distal to the point mutation. Also striking is the fact that two other mRNAs having a hypothetically different base substitution (A or G) at the same position assume a structure similar to that of RanT/n, emphasizing further that the single base change at this particular position is important.

View larger version (34K): [in this window] [in a new window]   Fig. 1.   Structure analysis of RanT/n and RanC/d mRNAs. A, computer analysis of the secondary structure of 3'-UTR Ran cDNAs, with all four possible substitutiona at position 870 of the Ran mRNA (the ATG start codon is position 1) based on the Squiggle program in the GCG computer analysis package. B, RNase T1 digestion. In vitro synthesized RNA molecules containing full-length 3'-UTRs of RanC/d or RanT/n mRNAs were subjected to RNase T1 digestion as described under "Materials and Methods." Labels d and n represent 3'-UTRs of RanC/d and RanT/n, respectively. Both RNase T1 dose and time course experiments showed distinctive band patterns of d and n, suggesting different tertiary structures of 3'-UTRs of the T/n and C/d mRNA.

To confirm the RNA structural difference predicted by computer analysis, we performed RNA digestion experiments where the three-dimensional RNA structure of RanT/n or RanC/d was examined. The mRNAs were digested with RNase T1, which targets guanines in single RNA strands (40). A 0.6-kilobase RNA fragment of either Ran cDNA, covering the 251-bp coding region and the whole 3'-UTR (344 bp from the stop codon to the beginning of poly(A)), was synthesized and digoxigenin-labeled in vitro. They were then digested with various amounts of RNase T1 for various lengths of time, and the digested fragments were fractionated by electrophoresis in a 10% urea-polyacrylamide electrophoresis gel. In the digestion reaction using RNase T1 at a concentration of 0.01 unit/µl, but not 0.1 unit/µl or more, we observed a distinct band in RanC/d samples (lanes labeled as d in Fig. 1) but not in RanT/n samples (lanes labeled as n in Fig. 1). The same band resistant to RNase T1 treatment was observed in RanC/d samples but not in RanT/n samples when the treatment time was extended from 5 to 45 min. These data are consistent with the presence of secondary and tertiary structural difference between RanT/n and RanC/d mRNAs.

Striking Difference in Steady-state Protein Amounts in RanT/n- and RanC/d-transduced Primary B Lymphocytes or Macrophages-- To assess whether this striking structural difference might be related to the biological differences between the two molecules we have observed to date (23, 29, 30), we designed more experiments to study their transcriptional and translational regulation. We are aware that Ran isoform genes exist in the mouse genome (41, 42). To differentiate exogenous and endogenous Ran protein, we tagged both exogenous Ran proteins by inserting a 39-bp HA epitope sequence into the Bsu36I site located close to the end of the coding sequence of the Ran cDNAs (Fig. 2). This design would allow us to quantify the amount of exogenous RNA and protein expressed by the Ran transgenes. To ensure the validity of our experimental results, we inserted both epitope-tagged cDNAs into both retrovirus and adenovirus vectors. The former will allow for stable incorporation of the transgenes into the genome of a variety of established cell lines, the latter for transient gene expression in primary macrophages and B lymphocytes from various strains of mice.

View larger version (23K): [in this window] [in a new window]   Fig. 2.   Vector design, maps, and structures. A, HA-tagged Ran cDNAs. HA tag was inserted into RanC/d and RanT/n cDNAs, respectively, into position between Glu186 and Val187 as described under "Materials and Methods." As a result of such an insertion, the HA-tagged Ran genes can be differentiated from endogenous Ran genes by PCR analysis using a pair of primers flanking the insertion site. PCR will generate a 119-bp fragment from endogenous Ran genes, and a 158-bp fragment from tagged Ran cDNAs with primers 5'-TTGTTGCCATGCCTGCTCTTG-3' and 5'-GGTCATCATCCTCATCTGGGA-3'. B, illustration of three recombinant plasmids used in the studies. HA-tagged Ran cDNAs, RanC/d-HA, and RanT/n-HA were constructed into a retroviral vector pN2, becoming pN2-RanC/d-HA and pN2-RanT/n-HA, respectively. C, construction of adenovirus vectors. HA-tagged Ran cDNAs were inserted into adenoviral vectors as described previously (29).

We transduced p388D1 macrophage cells (46) and 70Z/3 pre-B lymphocytes (23) by retroviral infection with N2-Ran retroviruses (with or without HA tag) and G418 selection (35). We extracted DNA from pools of more than 100 G418^r clones for each line and determined the DNA copy number in the transduced cells by Southern blot analysis. After BamHI digestion, the intensity of the 1.1-kb band (which indicates the intact Ran-HA gene) was comparable in all cells transduced with N2-RanT/n-HA or N2-RanC/d-HA vector, suggesting similar DNA copy number of the Ran transgene in all transduced cells (Fig. 3A). Digestion with BclI, which has only a single recognition site within the retroviral genome, releases individual retroviral junction fragments. In this regard, while the endogenous Ran bands were clearly present in all three lanes including the N2 control, no vector-associated band could be observed, which is consistent with the prediction that the genomic DNA were derived from a pool of more than a 100 distinct clones of retrovirally transduced cells.

View larger version (35K): [in this window] [in a new window]   Fig. 3.   Retroviral gene transfer and expression studies. A, Southern blot analysis of HA-tagged Ran gene transduction. Genomic DNAs were isolated from GPE packaging cells transduced with different retrovirus plasmid DNAs, or from P388D1 and 70Z/3 cells infected with retrovirus vectors N2, d, and n (representing the three retroviral plasmids in Fig. 2B). Genomic DNAs were digested with BamHI and BclI, which will release 1.1-kb Lps cDNA and flanking sequences, respectively. Electrophoresis, transfer, hybridization, and chemiluminescent detection were performed as described under "Materials and Methods." B, BamHI; L, BclI. B, RNase protection assay. The same transduced cells were used as in the Southern blot analysis. GTC cell lysates, DIG-labeled probes, Ran-HA, and actin, were prepared and RNase protection assays were carried out as described under "Materials and Methods." Probes and protected fragments were illustrated below the figure. As shown, protected actin fragment, HA-tagged Ran, and endogenous Ran are 245, 158, and 87 nucleotides in length, respectively. C, Western blot analysis. Protein expression levels of HA-tagged Ran were detected with anti-HA antibody (12CA5) from the same sets of transduced cells as in the Southern blot analysis. HA-tag specific protein bands were only detected in Ran-HA transduced cells, but not in control cells.

Next we measured the mRNA level at steady state by RNase protection assay using probes specific for Ran-HA and actin genes. A significant amount of Ran-HA mRNA (158 nt) was present in all Ran-transduced cells and was absent in all N2-transduced cells (Fig. 3B). At steady state, the intensity of the 158-nt Ran-HA band was not significantly different in GP/E fibroblasts, p388D1 macrophages, and 70Z/3 pre-B lymphocytes. Further time course RNase protection assays on actinomycin D-treated or 5,6-dichlorobenzimidazole riboside-treated Ran-transduced P388D macrophages, 70Z/3 B cells, and primary peritoneal macrophages from C3H/HeOuJ mice, suggested that the difference in stability between the two mRNAs was not significant (Fig. 4, not shown).

View larger version (15K): [in this window] [in a new window]   Fig. 4.   Half-life analysis of RanT/n mRNA and RanC/d mRNA. Two populations of retroviral transduced 70Z/3 cells (by N2-RanC/d-HA and N2-RanT/n-HA) were evaluated for the half-life of Ran-HA mRNA as described under "Materials and Methods." These cells were stimulated overnight with or without LPS at 1 µg/ml, and then treated with 100 µM 5,6-dichlorobenzimidazole riboside (DRB) for a period of 0, 2, 4, and 8 h, respectively. The amount of Ran-HA mRNA at each time point was analyzed and presented as % of mRNA remaining compared with that at time 0. The figure is the results of regression analysis from three independent experiments. The labels d and n represent 70Z/3 cells transduced with N2-RanC/d-HA or N2-RanT/n-HA. The labels "" and "+" represent without LPS or with LPS stimulation. The error bars are standard deviations derived from the mean values of the percentage of mRNA remaining at each time point.

We then measured the amount of protein derived from the transgenes by Western blot analysis. Surprising differences were observed. While the amounts of exogenous proteins from both transgenes were comparable in GP/E fibroblasts, the amounts of RanC/d protein was higher than that of RanT/n by 5-10-fold in both p388D1 macrophages and 70Z/3 pre-B lymphocytes (Fig. 3C). This striking difference was also present in primary peritoneal macrophages expressing the RanT/n and RanC/d transgenes after adenoviral gene transfer (73).

Similarly, we also studied primary B lymphocytes and macrophages transduced with Ad-Ran vectors. Fig. 5 shows that while the DNA copy number and the RNA copy number of the transgenes remained similar, the amount of exogenous Ran in these transduced cells appears to be strikingly different. Specifically, RanC/d-transduced cells have a 5-10-fold higher amount of the exogenous Ran protein than RanT/n-transduced cells. This difference appears to be relatively independent of viral infection, mouse strains, and LPS stimulation. Similar data were obtained in macrophages from C3H/HeOuJ and C3H/HeJ mice in that macrophages transduced with RanC/d-HA contained the protein at a level significantly higher than that of RanT/n-HA (30).

View larger version (40K): [in this window] [in a new window]   Fig. 5.   Transduction of splenic B cells with Ad-RanT/n-HA or Ad-RanC/d-HA. A, PCR analysis. Splenic B cells were infected with adenovirus carrying the RanT/n-HA (labeled as n) or the RanC/d-HA gene (labeled as d), at different multiplicity of infections ranging from 1:100 to 1:10,000 (labeled as 100, ... , 104). DNA was extracted from various samples and amplified using Ran-specific primers as described under "Materials and Methods." The 158-bp band indicates the presence of the transduced Ran-HA gene (labeled as HA-Ran), whereas the lower 119-bp band indicates the presence an endogenous Ran gene. Note the comparable intensity of the HA-Ran band in each paired DNA from Ad-infected B cells. B, RT-PCR analysis. Identical infection experiments, except RNA extraction, was performed and Ad-infected splenic B cells from C3H/HeOuJ and C3H/HeJ mice were analyzed. + or  signs indicate with or without 1 µg/ml LPS stimulation for 24 h. OuJ, C3H/HeOuJ; HeJ, C3H/HeJ. No, No infection; d, infection with Ad-RanC/d; n, infection with Ad-RanT/n. C, Western blot analysis. Cell lysates from uninfected, adenovirus-infected cell cultures, with or without 1 µg/ml LPS stimulation for 24 h were analyzed. The same filter was blotted with an antibody mixture containing mouse anti-tubulin antibody and rat anti-HA antibody, and the signal visualized as described under "Materials and Methods."

Differential Intracellular Distribution of the Exogenous Ran Protein in RanC/d- and RanT/n-transduced Macrophages-- Since the proteins from either RanC/d or RanT/n mRNAs are identical, the striking difference seen only at the protein level prompted us to examine if the exogenous RanT/n or RanC/d mRNAs or their proteins may have differential intracellular localizations. Adherent primary macrophages cultured onto plastic slides were infected with various Ran-HA adenovirus vectors, with or without LPS stimulation for 2 h, and then subjected to immunofluorescence analysis using a sandwich anti-HA antibody system described under "Materials and Methods." We observed three types of staining patterns (Fig. 6). For macrophages infected with Ad-RanC/d, more than 60% of the cells contained nuclear Ran, N; 20-30%, cytoplasmic Ran, C; and 10%, perinuclear membrane, NM. On the other hand, for macrophages infected with Ad-RanT/n, at the same 2-h time point, only 20-30% contained nuclear Ran, N; but 60-70%, cytoplasmic Ran, C; and 10%, perinuclear membrane Ran, NM. At the 24-h time point, most of the exogenous Ran was found in the nucleus in all cases.

View larger version (22K): [in this window] [in a new window]   Fig. 6.   Intracellular localization studies. A, immunofluorescent staining of HA-Ran proteins in macrophages transduced with either RanT/n or RanC/d gene. Primary macrophages from thioglycollate-primed C3H/HeOuJ or C3H/HeJ mice were infected with either Ad-RanC/d-HA (HA-Lpsd) or Ad-RanT/n-HA (HA-Lpsn), and stained with rat anti-HA antibody, followed by goat anti-rat IgG conjugated with biotin and streptavidin-fluorescein isothiocyanate. The slides were mounted with Vectashield mounting medium with propidium iodide (PI). Images of fluorescein isothiocyanate (top) and PI (bottom) were acquired using a confocal laser scanning microscope. Cell types with 3 distinct patterns of immunofluoresence were recorded. Type C indicates cells with a majority of cytoplasmic HA-Ran protein; type NM are cells with the HA-Ran protein localized externally around the nuclear membrane; and type N are cells whose nuclei stained positively with the anti-HA antibody. B, intracellular distribution of Ran protein at the 2-h point. After cell culture and adenoviral infection, with or without 1 µg/ml LPS for 2 h, 100 HA positively stained cells were randomly examined at high magnification and the types of positive cells as defined in A were recorded. The white color indicates the proportion of cells with cytoplasmic HA staining, C; gray shows the proportion of cells with staining around the nuclear membrane, NM; and black is the proportion of cells with nuclear staining. /+ indicates cultures with or without 1 µg/ml LPS. OuJ, macrophages from C3H/HeOuJ mice; HeJ, macrophages from C3H/HeJ mice. Results of one experiment were shown in the figure. Similar results were obtained in another four experiments performed. C, time course analysis on the intracellular distribution of Ran in RanT/n- and RanC/d-transduced macrophages. The proportion of each type of HA-positive cells was recorded. Based on the results shown in Fig. 2, the data for cell types NM and C were combined, as NM represents a minor proportion of cells at all time. Solid line indicates the proportion of N type; dotted line is the proportion of NM + C types. Results of one experiment were presented. Similar patterns were observed in another four such experiments performed.

Overall Reduction of RanT/n but Not RanC/d Protein Correlated with Its Cytoplasmic Localization-- Cytoplasmic proteases are known to play a role in LPS signal transduction in B lymphocytes and macrophages (63-69). In light of this, the reduced amounts of exogenous Ran proteins in RanT/n-transduced cells may be the result of the "lingering" of Ran in the cytoplasm, rendering themselves available to degradation by cytoplasmic proteases (Figs. 3, 5, and 6). This would predict that the initial amounts of Ran proteins derived from either RanT/n or RanC/d transgenes are the same. Only with time, those in RanT/n-transduced cells would be lower than those in RanC/d-transduced cells.

To investigate this possibility, we performed Western blot analysis on cell lysates at 2 or 24 h, with or without LPS stimulation. Indeed, at the 2-h point, the amount of Ran proteins were the same in RanT/n- or RanC/d-transduced macrophages (Fig. 7A). On the other hand, at the 24-h point, the amount of Ran was 5-10-fold lower in RanT/n-transduced macrophages than in RanC/d-transduced cells. The reduction of cytoplasmic RanT/n is also generally correlated with a more drastic decline in the total number of HA-positive cells as compared with that of RanC/d-transduced cells. Furthermore, the amount of Ran in the nuclei was found to be higher in cells transduced with RanC/d than that with RanT/n cDNA (not shown).

View larger version (28K): [in this window] [in a new window]   Fig. 7.   A, measurement of overall HA-Ran proteins. Primary macrophages were infected with Ad-RanT/n or Ad-RanC/d viruses, with or without 1 µg/ml LPS stimulation for 2 or 24 h. Five micrograms of total proteins of various cell lysates were loaded in each lane. After electrophoresis and blotting, the filter was incubated with two antibodies, mouse anti-tubulin and rat anti-HA; then with biotin-labeled goat anti-mouse IgG heavy chain and biotin-labeled goat anti-rat IgG heavy chain antibodies. Next, the filter was incubated with streptavidin-POD conjugate and visualized by addition of chemiluminescent substrate, luminal/iodophenol, according to the manufacturers instruction (see "Materials and Methods"). OuJM, macrophages from C3H/HeOuJ mice; HeJM, macrophages from C3H/HeJ mice; n or d, macrophages transduced with Ad-RanT/n or Ad-RanC/d adenovirus, respectively. B, measurement of cytosolic and nuclear HA-Rans. Nuclear and cytoplasmic extracts were prepared according to Welch and Wang (44). Ten micrograms of cell lysates were loaded in each lane and processed similarly as in A.

To investigate this further, we performed cell fractionation experiments and quantified the amount of HA-Ran proteins in cytosolic and nuclear compartments. Results in Fig. 7B show that at the 2-h time point, the amount of cytosolic HA-Ran in T/n-transduced cells was higher than those in C/d-transduced cells; whereas the amount of nuclear HA-Ran in C/d-transduced were higher than that in T/n-transduced cells. With time, the amount of cytosolic HA-Ran in both T/n- or C/d-transduced cells became minimal or undetectable, whereas the amounts of nuclear HA-Ran remained higher in C/d-transduced cells than in T/n-transduced cells. Thus, a fast accumulation of exogenous Ran in the nucleus is correlated with the maintenance of higher levels of Ran with time, and cytoplasmic Ran appears to be vulnerable to certain degradation process.

Also consistent with the idea of the presence of cytoplasmic proteases degrading Ran is the rate of degradation of both proteins when we determined the kinetics of exogenous Ran degradation by pulse-chase analysis. Primary macrophages were infected with the recombinant adenovectors, pulsed for 1 h with [³⁵S]methionine, and then chased for 0-24 h. After linear regression analysis on data from various time points, the protein from RanC/d mRNA appears more stable than that from RanT/n mRNA (Fig. 8). Most interestingly, the data also suggest the presence of two types of proteins in RanT/n-transduced cells: one gets degraded more rapidly than the other, whereas the curve for Ran in RanC/d-transduced cells appears to be a single species.

View larger version (18K): [in this window] [in a new window]   Fig. 8.   Degradation of Ran-HA proteins in transduced primary macrophages. Adenovirus-infected primary macrophages were pulsed with [35S]methionine for 1 h and chased for various lengths of time. The cells were then lysed and immunoprecipitated with anti-HA antibody, proteins resolved on gels, and the amount determined by fluorography. Error bars are standard deviations of the means.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Results from both computer analyses as well as gel retardation assay reveal a significance structural difference that exists between the RanT/n and RanC/d RNA molecules. This difference co-localizes with that of the point mutation at position 870 in the 3'-UTR of the gene, and in regions distal to this point mutation no significant structural difference is observed (Fig. 1). The difference is further emphasized by the finding that the theoretical substitution of a G or A residue at the same position reveals an identical structure as T/n but not C/d. This structural difference correlates with all the biological differences we have been observing in our gene transfer studies, in which the only difference in the adenovirus vectors we used is the single nucleotide base change in the Ran cDNA within the whole viral genome (29, 30).

This point mutation did not appear to affect the overall level of the mRNA nor its stability (Figs. 3 and 4). The translational rates of the two proteins also appear to be the same, as the initial amounts of the exogenous Ran were the same (Fig. 8). With time, the amounts of exogenous Ran in RanT/n-transduced cells were drastically reduced, whereas those in RanC/d cells changed only slightly (Fig. 8). This drastic reduction is correlated with its lingering in the cytoplasm (Fig. 7), and with the presence of two species of Ran in T/n- but not C/d-transduced cells, one of which probably resides in the cytoplasm and gets degraded rapidly, and the other resides in the nucleus (Fig. 9). This reduction further suggests its involvement in a cytoplasmic protein degradation process known to occur in LPS-induced signal transduction. LPS-mediated stimulation of B cells in mice was abrogated by treatment with the protease inhibitor, diisopropylfluorophosphate, and that induction of B cell proliferation correlated with enhanced expression of an intracellular arginine-specific serine enzyme (65). Trypsin synergizes with LPS in partially restoring the hypo-responsiveness of B cells from C3H/HeJ mice (66). Duc Dodon and Vogel (67) reported that LPS suppresses a secretory elastase activity from macrophages in LPS sensitive mice but not in LPS hyposensitive C3H/HeJ mice. By differential display, Jin et al. (68, 69) identified the known protease inhibitor SLPI as well as the known matrix metalloprotease-9, an active proteolytic enzyme; both the protease and the protease inhibitors are constitutively expressed in C3H/HeJ cells, and, although both are not highly expressed in cells of endotoxin-sensitive mice, both are inducible upon LPS stimulation. These studies reveal the presence of an active membrane-bound cytoplasmic protease system through which LPS signaling is mediated. Our protein localization study was also revealing. At the 2-h point, about 60% of Ran in T/n-transduced cells remained in the cytoplasm, whereas about 70% of Ran in C/d-transduced cells were in the nucleus (Fig. 3). On a more refined analysis, we noticed that, within the subpopulation of cells positive for cytoplasmic Ran, the proportion of cells with more intense perinuclear staining appeared higher in cells transduced with RanC/d than with RanT/n. Taken together, the difference in RNA structure observed in this study appears to dictate the intracellular distribution of the RanT/n and RanC/d proteins.

View larger version (14K): [in this window] [in a new window]   Fig. 9.   A model for Ran-mediated signal transduction. We hypothesize a 2-signal transduction mechanism. When Ran is in the cytoplasm, it transduces a positive response signal, probably indirectly by binding to cytoplasm-specific proteins. When it is in the nucleus, it transduces a negative response signal, probably by binding to nuclear-specific proteins. The outcome of the LPS response would be a net balance between the positive and negative signals. The point mutation at position 870, exemplified by T/n and C/d genes, affects mRNA intracellular localization and, subsequently, degradation of the Ran proteins in the cytoplasm. As a result of its more rapid accumulation of Ran from RanC/d mRNA into the nucleus, a negative signal ensues in RanC/d-transduced cells or a refractory response when most of the Ran proteins in any Ran-transduced cells migrate into the nucleus.

The studies in this report also reveal the presence of a novel motif in the 3'-UTR of Ran mRNAs, which regulates the distribution of Ran proteins, affecting their overall amounts, and modulating the biological processes in which Ran proteins participate. The role of the subcellular RNA location in regulating biological processes has received increasing emphasis lately (49). There are several well documented examples in which genetic motifs located in the 3'-UTR of mRNAs regulate RNA subcellular localization, especially in regions rich in secondary structures (49). Kislauskis and Singer (50) call these motifs "zipcodes." During Drosophila development, the bicoid mRNA and therefore its protein is produced at the anterior pole of young embryos. There are two such zipcode signals, each resides within a stem loop structure located within the bicoid 3'-UTR (51). A number of proteins, which include Staufen and Exl, have been shown to bind to either one of the two bicoid RNA localization signals (49, 52, 53). In Xenopus, distinct zipcode motifs, as short as 57 nt, have been identified in 3'-UTRs of Vg1, Xcat, and Veg T mRNAs (54-56). In chicken fibroblasts, a 54-nt zipcode signal, which is recognized by the zipcode-binding proteins, ZBP-1 and ZBP-2, is present in the 3'-UTR of -actin mRNA (49, 57). The binding of ZBP-1 to -actin mRNA targets the RNA to the leading edge of motile fibroblasts (58). In neurons, protein binding to -actin mRNA results in RNA targeting to the growth cones of both immature dendrites and axons (59).

Ran has been shown to have diverse biological functions, such as nuclear transport, mitotic spindle formation, RNA modification, chromosome stabilization, and cell division progression (24-28). These diverse functions may be related to the fact that both cytoplasmic GDP-bound Ran and nuclear GTP-bound Ran may have distinct biological functions (24, 48). Here we propose a model for a two-signaling system in which the ratio of nuclear and cytoplasmic Ran dictates the outcome of LPS response in one or more signaling pathways. As shown in Fig. 9, a low N/C ratio would transmit a positive LPS response; a high N/C ratio would result in a negative response. During a normal acute phase of LPS response, a higher level of cytoplasmic RanT/n transmits a positive response signal. With time, reduction of cytoplasmic Ran occurs, via its migration into the nucleus, its degradation by cytoplasmic proteases, or both, would confer a state of refractoriness to further LPS stimulation. This autocrine negative feedback regulation is consistent with the finding by Fahmi and Chaby (70), who showed that the refractoriness of LPS-induced TNF production by macrophages is via an autocrine mechanism. Another demonstration of autocrine reduction of TNF production is the endogenous feedback activation of tritetraprolin, which has been shown to bind to the AU-rich element in the 3'-UTR of TNF mRNA, destabilizing the messenger RNA and therefore reduction of the cytokine production (71). Thus the N/C ratio of Ran would represent the net-balanced activities of both the intracellular proteases and protease inhibitors.

Ran may be involved in more than one host inflammatory signal transduction pathway (60-62), the tissue-specific difference observed in macrophages and B lymphocytes but not fibroblasts might indicate that a "preferred" signaling pathway may be more dominant in one cell type but not in another (Figs. 3 and 8). In this light, differences between macrophages and B lymphocytes also exist. RanT/n is less effective in restoring an LPS response in C3H/HeJ macrophages (10-20%) than in C3H/HeJ B lymphocytes (35-50%) (23, 29). Given the recent exciting findings that a number of membrane-bound toll receptors are involved in innate immune responses (10-22, 63, 64), and as Ran is an intracellular protein whose distribution is primarily nuclear at steady state, it may act downstream in multiple signal transduction pathways in innate immune responses initiated by various Tlr receptors at the cell surface.

Tlr4 gene has been mapped to the Lps locus and tlr4-d has a missense mutation in the genome of C3H/HeJ mice resistant to endotoxin challenge (8, 9). Tlr4 knockout mice are clearly hyporesponsive to LPS challenge, as do mice with homozygous deletion of MyD88 and CD14 genes (22, 60, 72). Thus Tlr4 is clearly an Lps gene (8, 9). Whether or not Tlr4-d is the only Lps-defective gene in the genome of C3H/HeJ mice has not been shown unambiguously. In fact, two early studies suggest that the genetic basis for C3H/HeJ hyporesponsiveness could be due to multiple genetic defects (17, 43). Furthermore, if Tlr4 were the only single mutated Lps gene in C3H/HeJ-resistant mice, one would predict a powerful phenotypic conversion to endotoxin resistance after adenoviral transfer of the Tlr4 cDNA into endotoxin-sensitive mice. Such a demonstration has not yet been reported since its initial publication in 1998 (8). Is RanC/d a defective gene in the C3H/HeJ genome? We could not answer this question with certainty at this time for the following reasons. First, Ran has been shown by D'Eustachio and associates to have at least seven isoforms that are highly homologous in the mouse genome (41, 42), and by our collaborators during our in situ hybridization studies.^2,3,4 Second, we have shown that some of the Ran isoform genes in the mouse genome are intronless.⁵ Third, by SSCP-PCR analysis, we have shown that as many as five such Ran isoform genes have a T (thymidine) base at position 870 at the 3'-UTR of the gene in the mouse genome.⁵ Thus RanC/d might have been derived from a natural mutation in the C3H/HeJ genome, which we could not ascertain; alternatively, it might have derived from reverse transcription during cDNA library construction using mRNA derived from C3H/HeJ splenic B cells (73). Regardless of its origin, the results in this study further strengthen our previous finding that Ran is involved in endotoxin LPS signal transduction; this finding was established based on the powerful approach of functional cDNA expression cloning (35). Together with the use of many different types of functional assays employed to compare the biological properties of RanT/n- and RanC/d-transduced cells and mice, the results in this study once again illuminates the presence of a novel motif at the 3'-UTR of Ran regulating all the reported biological changes of Ran in a unique way as a result of a single nucleotide base substitution (23, 29, 30, 62, 73).

	ACKNOWLEDGEMENTS

We thank Drs. Y. W. Kan, Fangping Zhou, Pei-dong Fan, Rachel Sheppard, HongYuan Luo, and James Heuck for helpful discussions or technical assistance.

	FOOTNOTES

^* This work was supported by National Institutes of Health Grants RO1CA70854, RO1AI39159, and RO1AI45951 (to P. M. C. W.).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: Dept. of Pathology & Laboratory Medicine, Fels Institute, Temple University School of Medicine, 3307 North Broad St., AHB, Rm. 552, Philadelphia, PA 19140. Tel.: 215-707-8361; Fax: 215-707-8351; E-mail: Petermcw@aol.com.

Published, JBC Papers in Press, June 27, 2001, DOI 10.1074/jbc.M105400200

² N. Jenkins, personal communication.

³ H. Heng, personal communication.

⁴ T. R. Chen, personal communication.

⁵ P. M. C. Wong, Q. Yuan, H. Chen, B. M. Sultzer, and S-W. Chung, manuscript in preparation.

	ABBREVIATIONS

The abbreviations used are: LPS, lipopolysaccharide; UTR, untranslated region; RanT/n, Ran thymidine/normal; RanC/d, Ran cytidine/deficient; TNF, tumor necrosis factor ; HA, hemagglutinin; PCR, polymerase chain reaction; bp, base pair(s); nt, nucleotide(s); DIG, digoxigenin; PBS, phosphate-buffered saline.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articl