Am J Physiol Cell Physiol 289: C264-C276, 2005.
First published March 2, 2005; doi:10.1152/ajpcell.00461.2004
0363-6143/05 $8.00
RECEPTORS AND SIGNAL TRANSDUCTION
Mechanism of amyloid peptide induced CCR5 expression in monocytes and its inhibition by siRNA for Egr-1
Ranjit K. Giri,1
Vikram Rajagopal,1
Shweta Shahi,1
Berislav V. Zlokovic,2 and
Vijay K. Kalra1
Department of 1Biochemistry & Molecular Biology, Keck School of Medicine, University of Southern California, Los Angeles, California; and 2Center for Aging, University of Rochester, Rochester, New York
Submitted 20 September 2004
; accepted in final form 16 February 2005
 |
ABSTRACT
|
|---|
In Alzheimer's disease (AD), one finds increased presence of monocytes/macrophages and activated microglial cells in the brain. Immunohistochemical studies show increased expression of chemokine receptor 5 (CCR5) on reactive microglia associated with amyloid deposits in AD, suggesting that CCR5 may play a role in the regulation of the immune response in AD. In this study, we used peripheral blood monocytes and human monocytic THP-1 cell line as a model of microglia to delineate the cellular signaling mechanism of A
-induced CCR5 expression and the latter's role in the chemotaxis of monocytes. We observed that A peptides at pathophysiological concentrations (125 nM) increased CCR5 mRNA and cell surface protein expression. The cellular signaling involved activation of c-Raf, ERK-1/ERK-2, and c-Jun NH2-terminal kinase. Analysis of some transcription factors associated with CCR5 promoter revealed that A increased DNA binding activity of Egr-1 and AP-1. In addition, we show that CCR5 promoter contains an Egr-1 like consensus sequence GCGGGGGTG as demonstrated by 1) electrophoretic mobility shift assay, 2) transfection studies with truncated CCR5 gene promoter construct, and 3) chromatin immunoprecipitation analysis. Moreover, transfection of Egr-1 siRNA, but not of scrambled Egr-1 siRNA, in THP-1 cells resulted in >75% reduction in both A-mediated CCR5 expression and concomitant chemotaxis to its ligands. We suggest that inhibition of Egr-1 by either Egr-1 siRNA or pharmacological agents may reduce activation of monocytes/microglia and possibly ameliorate the inflammation and progression of AD.
amyloid peptide; chemokine receptor 5; small inhibitory ribonucleic acid
IN ALZHEIMER'S DISEASE (AD) and amyloid peptide (A)-related cerebral vascular disorders (cerebral amyloid angiopathy and hereditary cerebral hemorrhage with amyloidosis of Dutch type), one finds increased deposition of A in both the brain parenchyma and cerebral vasculature (30, 33). There is growing evidence that AD is a neuroinflammatory disease characterized by the increased presence of activated microglial cells and astrocytes in the brain (19, 20, 32). Both of these cells generate inflammatory mediators, such as complement proteins, cytokines, and chemokines in response to the A peptide (20). Because peripheral blood monocytes can cross the blood-brain barrier (BBB) and undergo differentiation into microglial cells (5, 12), we sought to determine the molecular basis of how peripheral blood monocytes/macrophages may accumulate in the AD brain in response to increased presence of amyloid peptide in circulation and in the brain of AD patients. In our previous studies (8), we showed that both the soluble (A1–40) and fibrilar (A1–42) form of amyloid peptides could induce the transmigration of human monocytes across cultured brain endothelial cell monolayer, a model of BBB. However, it is unclear whether A-activated monocytes could migrate across the cerebral vasculature as a result of a chemotactic gradient generated by chemokines, which are elaborated from activated microglial cells and astrocytes in the AD brain (21).
Immunohistochemical studies (32) show increased presence of chemokine macrophage inflammatory protein (MIP)-1 in a subpopulation of reactive astrocytes and MIP-1
in neurons of AD brain than those of controls. Moreover, their studies (32) showed that chemokine receptor (CCR)3 and -5 are present on microglial cells of both control and AD brain but with an increased expression on reactive microglia in AD. Because CCR5 is expressed on monocytes and certain lymphocytes, and is activated by the -chemokines [MIP-1, MIP-1, and regulated on activation normal T-expressed and presumably secreted (RANTES)] (25), we hypothesized that the increased expression of CCR5 on microglial cells or their precursors (monocytes/macrophages) could occur as a result of activation by A peptides. Moreover, these activated monocytes may transmigrate across the brain vasculature in response to -chemokines (MIP-1, MIP-1, and RANTES) elaborated by microglia and astrocytes in the brain.
The role of CCR5 and its ligands in human immunodeficiency virus (HIV) infection has been the subject of much scrutiny (1, 23, 24). Studies (1, 28) have shown that CCR5 serves as the main coreceptor for the entry of HIV in monocytes/macrophages and humans having a 32-base pair (bp) mutation in the CCR5 gene (CCR5–32 mutant allele) are generally resistance to HIV-1 infection. In vitro studies of peripheral blood lymphocytes show decreased cell surface expression of CCR5 in response to TNF- (13) and LPS (7). However, relatively less is known of the mechanism by which amyloid peptides induce the expression of CCR5 in monocytes/macrophages, the progenitor cells of microglia.
In the present study, we show that A peptides (A1–40 and A1–42) at pathophysiological concentrations (125 nM), as found in the plasma of AD individuals (14), cause cellular signaling in THP-1 monocytic cells and peripheral blood monocytes to increase the gene expression of CCR5. The cellular signaling for increased expression of CCR5 in response to A involves activation of Raf kinase and MAP kinase (ERK1/ERK2). We (10) recently showed that A at submicromolar concentration causes activation of transcription factor AP-1 and Egr-1 but not of NF-
B and cAMP response element binding protein (CREB) in THP-1 monocytes and peripheral blood monocytes (PBM). The presence of a putative Egr-I consensus sequence in the CCR5 promoter indicated that this transcription factor could regulate expression of CCR5. In the present study, we show for the first time by EMSA, transfection of THP-1 cells with truncated CCR5 gene promoter construct, and chromatin immunoprecipitation analysis that Egr-1 binds in vitro and in vivo to the newly identified consensus sequence of the CCR5 promoter. Moreover, transfection with small inhibitory RNA (siRNA) for Egr-1 mRNA abrogates A-induced CCR5 expression. In addition, we show that A-induced CCR5 expression in THP-1 monocytes plays a role in chemotaxis in response to -chemokines (MIP-1 and RANTES) as well as to amyloid peptide. We demonstrate that transfection of THP-1 monocytes with Egr-1 siRNA abrogates chemotaxis in response to MIP-1. For the first time, these studies demonstrate the importance of Egr-1 transcription factor in the regulation of CCR5 expression in monocytes.
|
METHODS
|
|---|
Amyloid peptides and their fibrillation state.
Human amyloid peptides, A1–40 and A1–42, were custom synthesized and characterized (10). Both the peptides were dissolved in endotoxin-free water (Sigma, St. Louis, MO) at a concentration of 2 mg/ml. We determined the fibrillar formation in these peptides preparations at regular time intervals during storage using a thioflavin T fluorescence assay (11) and far-ultraviolet CD spectra. A1–40, when freshly prepared in water, was monomeric, although it showed a small amount (
10%) of dimeric form when kept for 7 days, as analyzed by electrophoresis on native gel, followed by Western blot analysis with an antibody to A1–40. However, A1–42 (2 mg/ml), when dissolved in water and kept at 37°C for 7 days, showed fibrillar content. Both peptide solutions were negative for endotoxin (<10 pg/ml), as determined by limulus lysate test (Sigma) (10).
Reagents.
PD-98059, U-0126, and genistein were purchased from BIOMOL (Plymouth Meeting, PA). GW-5074, SB-203580, and SP-600125 were obtained from Tocris Cookson (Ellisville, MO). Rabbit anti-Egr-1 (SC-110X) and goat anti-SP-1 (SC-59X) were purchased from Santa Cruz Biotechnology (Santa Cruz, CA). Monoclonal antibody to CCR5, goat anti-MIP-1 and RANTES were obtained courtesy of NIH AIDS Reagent and Reference Program, operated by McKesson Bioservices (Rockville, MD). Recombinant human MIP-1 and goat anti-MIP-1 were obtained from R&D Systems (Minneapolis, MN). All other reagents, unless otherwise specified, were purchased from Sigma.
Cell culture and isolation of peripheral blood monocytes.
The THP-1 promonocytic cell line obtained from ATCC (Manassas, VA) was cultured in RPMI 1640 containing 10% heat-inactivated fetal calf serum, as described earlier (10). On the day of the experiment, THP-1 cells (1 x 106 cells/ml) were cultured in serum-free RPMI 1640 for 4–6 h. PBMs were isolated from blood collected in EDTA as the anticoagulant, as previously described (10).
RNase protection assay.
THP-1 monocytes were treated with amyloid peptides for various time periods and total RNA was isolated with TriZOL reagent (Invitrogen, Carlsbad, CA). RNase protection assays (RPA) were performed using custom made multiprobe templates for CCR5, CCR2a, and CCR2b and the housekeeping genes L32 and GAPDH (Pharmingen, San Diego, CA), as previously described (10). The intensity of bands corresponding to CCR5, CCR2a, CCR2b, and GAPDH were analyzed using a gel documentation system (model 2000, Alpha Imager, San Leandro, CA). Values were expressed as relative expression of mRNA normalized to the means of L32 and GAPDH mRNA values.
EMSA for transcription factors Egr-1.
THP-1 cells (5.0 x 106 cells) were treated with A1–40 (125 nM) for varying time periods (15–240 min) and nuclear extracts were prepared as described previously (10). The oligonucleotide probes used for CCR5-Egr-1 (putative Egr-1 binding site in CCR5 promoter) were as follows: 5'-(GTC CCT ATA TGG GGC GGG GGT GGG GGT GTC T)-3' and 3'-(CAG GGA TAT ACC CCG CCC CCA CCC CCA CAG A)-5', which were synthesized at Norris Cancer Center Microchemical core facility of University of Southern California-Keck School of Medicine (Los Angeles, CA). Probes were 5' end labeled with 100 µCi of [
-32P] ATP using T4-polynucleotide kinase (10). The DNA binding reaction mixture contained nuclear proteins (2–4 µg), [32P]-labeled double-stranded oligonucleotide probe (50,000 cpm), and 2 µg of poly dI-dC. To demonstrate specificity of DNA-protein interaction, 50-fold excess of unlabeled double-stranded oligonucleotide probe was added to the nuclear extract before the addition of radiolabeled probe. In supershift assays, nuclear extracts were preincubated for 20 min at room temperature with 2 µg of antibody specific to each transcription factor, before the addition of radiolabeled probe. The DNA-protein complex was then size fractionated from the free DNA probe by electrophoresis in a 4% nondenaturing polyacrylamide gel. The gel was dried and exposed to X-ray film.
Transient transfection of THP-1 cells and luciferase activity assay.
The firefly luciferase reporter gene plasmids of CCR5 promoter (PA-3; –731 to +33 regions) employed was kindly provided by Dr. Sunil Ahuja (San Antonio, TX). Mummidi and coworkers (23) have previously described their preparation and features. THP-1 cells (2–3 x 106 cells/well) were cultivated in 6-well chambers. The reporter gene constructs were transiently transfected in THP-1 cells by using Lipofectamine reagent (Invitrogen, Carlsbad, CA). Transfection efficiency was normalized by cotransfecting THP-1 cells with CCR-5 promoter-luciferase constructs (10 µg/well) and 0.5 µg of Renilla luciferase vector (pRL-CMV; obtained from Promega, Madison, WI). Alternatively, THP-1 cells cotransfected with 10 µg of the promoter less vector pGL3-Basic (Promega) and 0.5 µg of pRL-CMV was used as a negative control. After 2 days of transfection, the cells were pelleted, washed in Dulbecco's PBS, and lysed in 1x passive lysis buffer (Promega). The protein concentration in the cell lysates was determined by the Bradford method. The firefly and Renilla luciferase activities in the lysates were determined according to the manufacturer's instructions (Dual-Luciferase Reporter Assay System, Promega) utilizing a luminometer (Berthold Technologies, Oakridge, TN). The relative luciferase activity in each sample was determined as follows: X = firefly luciferase activity of CCR-5 promoter construct ÷ Renilla luciferase activity of pRL-CMV construct; Y = firefly luciferase activity of promoter less vector pGL3-basic ÷ Renilla luciferase activity of pRL-CMV vector; Z = X ÷ Y and relative luciferase activity is expressed as Z ÷ micrograms of protein in the lysate sample.
Chromatin immunoprecipitation assay.
THP-1 cells (5 x 106 cells) were serum starved for 6 h, followed by treatment with A1–40 for the indicated time period. Chromatin immunoprecipitation assay (ChIP) analysis was performed as described previously (26). Briefly, after stimulation of cells with A, cells were washed with PBS and then cross-linked with 1% formaldehyde at room temperature for 10 min. Cells were lysed, sonicated, and supernatants were then recovered by centrifugation of lysate at 12,000 rpm for 10 min at 4°C. The supernatant was diluted fourfold in a dilution buffer (1% Triton X-100, 2 mM EDTA, 150 mM NaCl and 20mM Tris·HCl, pH 8.1), followed by the addition of 2 µg of sheared salmon sperm DNA, 2.5 µg of preimmune serum, and 20 µl of protein A-Sepharose (50% slurry). The contents were kept at 4°C for 2 h. The precleared supernatant was immunoprecipitated by adding antibody (2 µg/ml) to either Egr-1 or SP-1, 2 µg of sheared salmon-sperm DNA, and 20 µl of protein A-Sepharose (50% slurry) and incubated at 4°C for 12–16 h. After several washings, the protein was digested with proteinase K (10 µg/ml) for 1 h. The cross-linking between DNA and protein was reversed by incubating the immunoprecipitate at 65°C overnight. DNA was phenol-chloroform extracted, ethanol precipitated, air dried, and dissolved in 50 µl of buffer composed of 10 mM Tris·HCl, pH 8.0, and 1 mM EDTA. A DNA sample (5 µl) was subjected to PCR amplification utilizing primers (5'-CCA GCA GCA TGA CTG CAG TT-3', forward primer; 5'-GCT AAT TGC TGG TGC TTG GAG-3' reverse primer) corresponding to the promoter region of CCR-5 (from –847 to –603, respective to the transcription start site).
Flow cytometry analysis.
THP-1 cells (5 x 106 cells) were incubated with A1–40 for indicated time period (1–4 h). Cells were collected, washed with ice-cold PBS, and resuspended in PBS at a concentration of 1 x 106 cells/ml. An antibody (5 µg/ml) was added to these cells to either CCR5 or CCR2b, and the contents were incubated at 4°C for 60 min. Cells were washed in ice-cold PBS, followed by incubation with FITC-conjugated goat antimouse IgG. These cells were washed with PBS and fixed at room temperature with 2% paraformaldehyde for 15 min. Fixed cells were washed three times in PBS and analyzed for CCR5 surface expression by flow cytometry. Gated acquisition of monocytes (10,000 events) was performed based on forward and side-scatter parameters.
Synthesis of siRNA duplexes for Egr-1 mRNA.
The 22 nucleotide sequence of Egr-1 siRNA was derived from human Egr-1 mRNA sequence (Genebank Accession No. GI: 5420378) and was targeted to the coding region 1237–1258 relative to the start codon of Egr-1 gene. Egr-1 siRNA and scrambled Egr-1 siRNA (scEgr-1 siRNA) were synthesized as previously described (10).
Transient transfection of THP-1 cells with Egr-1 siRNA duplex.
THP-1 cells were transfected with Egr-1 siRNA duplex utilizing lipofectamine (10). Briefly, Egr-1 siRNA or scEgr-1 siRNA (0.25 or 0.5 µg/ml) was incubated in 100 µl of serum-free DMEM containing 10 µl of lipofectamine (Invitrogen) for 15 min, followed by the addition to THP-1 cells. After 48–72 h of transfection, cells were harvested and used for further experiments.
Chemotaxis assay.
Chemotaxis was assayed in 96-well plates (Neuro Probe, Gaithersburg, MD) with Transwell inserts of 5-µm pore size. Briefly, THP-1 monocytes were washed and resuspended in serum free RPMI-1640 medium and 1 x 105 cells/50 µl were then loaded into an insert of the Boyden chamber. Chemotaxis medium (30 µl of serum-free RPMI-1640 medium) containing indicated amounts of chemokines or A was placed in the bottom compartment. After 2 h of incubation at 37°C in a 5% CO2 incubator, cells were scraped from the top chamber and washed with PBS (100 µl) to remove nonmigrated cells. This was followed by the addition of PBS containing 2 mM EDTA to the top chamber and incubation at 4°C for 15 min. Cells that had migrated into the lower compartment of the Boyden chamber were counted in five microscopic high-power fields (x40) with the use of an Olympus IMT-2 microscope. Where indicated, THP-1 cells were pretreated with A1–40 for 4 h, followed by a wash with serum-free medium, and used directly in the chemotaxis assay. In addition, the effect of neutralizing antibody against chemokines was determined by adding antibody to the chemoattractant protein in the lower compartment of the chemotaxis chamber. Where indicated, A-treated THP-1 cells were preincubated with antibody to chemokine receptor (CCR5 and CCR2b) for 1 h at room temperature. Each sample was tested in triplicate.
Statistical analysis.
Statistical analysis of the responses obtained from control and A-treated monocytic cells was carried out by one-way ANOVA utilizing Instat 2 (GraphPad, San Diego, CA) software program. The effects of inhibitors on A-induced responses were analyzed by comparing the response of THP-1 cells in the presence and absence of inhibitor. Dunnett's test was used for multiple comparisons. P values <0.05 were considered as significant.
|
RESULTS
|
|---|
A-induces CCR5 mRNA expression in THP-1 monocytes.
Because we observed that A induced expression of MIP-1 in THP-1 monocytes (10), we determined whether A concomitantly affected expression of its cognate receptor CCR5, which may participate in chemotaxis. As shown in Fig. 1A, both A1–40 and A1–42 at concentrations of 125–1,000 nM caused a dose-dependent increase in CCR5 mRNA, as determined by RPA analysis. At a dose of 125 nM A1–40 there was a twofold increase in mRNA expression of CCR5, which remained unchanged at a higher concentration (250–1,000 nM) of A (Fig. 1, A and B). However, for purposes of comparison, LPS at 100 ng/ml caused a fourfold increase in CCR5 mRNA expression (data not shown), which is contrary to the published studies seen in lymphocytes (7). Figure 1, C and D, shows the time course (1–4 h) of expression of CCR5 in response to A1–40 and A1–42, respectively. At each time point, we used untreated THP-1 cells as a control for CCR5 expression, as the baseline expression of CCR5 decreased with time. It is pertinent to note that both A1–40 (Fig. 1C) and A1–42 (Fig. 1D) did not affect the expression of CCR2b in THP-1 monocytes compared with control. The increase in mRNA expression of CCR5 was not due to contamination of A with endotoxin, as A was dissolved in endotoxin-free water. In addition, Polymyxin B (5 µg/ml), an inhibitor of endotoxin, did not significantly affect the A-induced mRNA expression of CCR5 (data not shown). The increase in the mRNA expression of CCR5 with 125 nM of A1–40 was optimal at 1 h of incubation (Fig. 1C). Similar results on CCR5 mRNA expression were observed with the fibrillar form of amyloid peptide, i.e., A1–42 (Fig. 1D). Because both soluble (A1–40) and fibrillar (A1–42) forms of amyloid peptide induced to the same extent the expression of CCR5, we used A1–40 in most of the studies described herein.
Role of tyrosine kinase, c-Raf kinase, MEK1/2, c-jun NH2-terminal kinase, and p38 MAP kinase in A-induced CCR5 mRNA expression.
Because A treatment of THP-1 monocytes increased mRNA expression of CCR5, we examined the effect of pharmacological inhibitors (2), which have been previously (10) shown to be specific for various kinases of the A-mediated signaling pathway. As shown in Fig. 2, pretreatment of THP-1 cells with genistein (25 µg/ml), a protein tyrosine kinase inhibitor, inhibited A-induced mRNA expression of CCR5 by <90%. A specific inhibitor of mitogen-activated protein kinase kinase (MAPKK/MEK), PD-98059 (10 µM), completely inhibited A-induced mRNA expression of CCR5 (P < 0.001). Similarly, U-0126 (150 nM), a specific inhibitor of MEK1/2, abrogated A-induced CCR5 mRNA expression. However, SB-203580 (1–2 µM), a selective p38 MAP kinase inhibitor, increased the A-induced expression of CCR5 mRNA (P < 0.001). The role of p38 MAP kinase remains to be elucidated. Next, we examined the effect of SP-600125 (100 nM), a potent inhibitor of c-Jun NH2-terminal kinase (JNK) (10). As shown in Fig. 2A, lane 7, SP-600125 reduced (80%) the expression of CCR5 mRNA. Moreover, GW-5074 (20 nM), a potent and specific inhibitor of c-Raf kinase (15), completely inhibited A-induced expression of CCR5 (Fig. 2A, lane 6). Although pharmacological inhibitors utilized are specific for these kinases (2), they may have other effects. These results suggest that A-induced mRNA expression of CCR5 involves activation of c-Raf kinase, MAPKK/MEK, and c-Jun NH2 terminal kinase.
A1–42 induced CCR5 mRNA expression in peripheral blood monocytes.
We determined whether interaction of amyloid peptide with PBM exhibited similar changes in CCR5 mRNA expression as observed in THP-1 monocytic cells. As shown in Fig. 3A, treatment of PBM with A1–42 (125 nM) resulted in an increase in CCR5 mRNA expression as was observed in THP-1 monocytes. Moreover, the pharmacological inhibitors PD-98059 (MEK-1/2 kinase inhibitor), GW-5074 (c-Raf kinase inhibitor), and SP-600125 (c-Jun NH2 terminal kinase inhibitor) reduced A1–42 mediated CCR5 expression in PBM (Fig. 3A). Similarly, A1–40 (125 nM) caused in an increase in CCR5 mRNA expression in PBM, which was reduced to the basal level in the presence of PD-98059 (Fig. 3B). It is pertinent to note in these PBM samples (Fig. 3B), the basal level of CCR5 was higher than seen in samples of PBM in Fig. 3A because blood samples of different donors were used. However, both A1–42 and A1–40 caused increases in CCR5 mRNA expression in PBM, which was almost completely inhibited by PD-98059. Because amyloid peptide showed similar profile of CCR5 mRNA induction in both THP-1 cells and PBM, we utilized THP-1 monocytic cells, for ease of culturing cells in needed quantity, as a model system of monocytes for subsequent studies.
Analysis of 5'-flanking region of CCR5 gene shows a functional Egr-1 binding site.
Our previous studies (10) showed that A activated transcription factor Egr-1 in THP-1 cells, as determined by EMSA. In these studies, we utilized oligonucleotide probe [upper strand, 5'-(GGATCCAGCGGGGGCGAGCGGGGGCGA)-3'] as the bona fide Egr-1 consensus sequence for EMSA analysis. We searched the human CCR5 promoter region in Genbank (Accession No. U95626) and sequence published by Mummidi and coworkers (23) for Egr-1 DNA binding sites. A search for potential transcription factor-binding sites was performed using the Transcription Element String Search (http://www.cbil.Lupenn.edu/tess), an Internet-based search engine. The analysis revealed the presence of cis-acting elements within 1.9 KB, upstream of the transcription start site. The data shows that human CCR5 promoter (23) contains GCGGGGGTG at positions –702 to –694. It is pertinent to note that the macrophage colony stimulating factor (M-CSF) promoter has a GCGGGGGAG sequence at positions –273 to –265 that is an Egr-1 binding site (29). On the basis of these observations, we hypothesized that the CCR-5 promoter with the GCGGGGGTG motif could be a DNA binding site for Egr-1. We utilized oligonucleotides [upper strand, 5'-(GTCCCTATATGGGGCGGGGGTGGGGGTGTCT)-3'] as the putative Egr-1 consensus sequence in CCR5 promoter (–715 to –685) for EMSA analysis. As shown in Fig. 4A, A-caused a time-dependent (0.25–2 h) increase in Egr-1 DNA binding activity utilizing CCR5-Egr-1 (GCGGGGGTG) sequence as a probe. Moreover, excess cold CCR5-Egr-1 probe completely abrogated the Egr-1 bands (1–3) in EMSA (Fig. 4A, lane 6). However, excess cold bona fide Egr-1 probe (GCGGGGGCG) only eliminated bands 2 and 3 (Fig. 4A, lane 7). Furthermore, antibody to Egr-1 caused a super shift of Egr-1 band (3) (Fig. 4A, lane 8), though control antibody to SP-1 did not cause super shift of any of these bands 1–3 (Fig. 4A, lane 9). These results indicate that Egr-1 DNA binding activity, corresponding to band 3, is likely to be involved in stimulating CCR5 promoter activity.
View larger version (38K):
[in this window]
[in a new window]
|
Fig. 4. CCR5 promoter analysis by EMSA and transfection studies with truncated CCR5 plasmid coupled to luciferase reporter. A: THP-1 cells were treated with either A1–40 (125 nM) for various time periods (0.25–2 h). Nuclear extracts were prepared and incubated with [32P]-labeled oligonucleotide probe for putative Egr-1 in CCR5 promoter. Where indicated, a 50-fold excess of unlabeled probe, antibody to Egr-1 (2 µg), and antibody to SP-1 (2 µg) were added to the nuclear extract. As shown in Fig. 1A, bands 1–3 appeared; however, antibody to Egr-1 caused super shift (SS) of only band 3. Thus band 3 corresponds to the Egr-1-DNA complex. The data are representative of three independent experiments. NS, nonspecific band. B: CCR-5 promoter constructs (PA-3) and pCMV Renilla luciferase construct were cotransfected into THP-1 monocytes. After 2 days posttransfection, cells were washed with serum-free media and either untreated or treated with A1–40 (125 nM) for 4 h. Where indicated, transfected cells were preincubated with PD-98059 (10 µM), SB-203580 (1 µM), and SP-600125 (100 nM), and then treated with A1–40. Cells were pelleted and washed once with PBS. Cell lysates were prepared and luciferase activity was measured by Dual luciferase assay kit (see METHODS). Results are expressed as the percentage of luciferase activity relative to untreated cells (n = 3, means ± SE). NS, not significant. ***P < 0.001, A1–40 treated vs. A1–40 treated in the presence of inhibitors.
|
|
A induces CCR5 promoter activity in THP-1 monocytes.
To determine whether A induces CCR5 mRNA expression at the transcriptional level, THP-1 cells were transfected with the luciferase reporter construct containing the CCR5 promoter region coupled to the 5' end of the luciferase reporter gene, designated as PA-3 (kindly provided by Dr. Sunil Ahuja, San Antonio, TX) (23). Previous studies (23) have shown that this region (–731 to +33) of CCR5 promoter in PA-3 construct contains cis acting elements for transcription factors SP-1 and putative Egr-1 binding site. As shown in Fig. 4B, transfection of THP-1 cells with PA-3 construct, followed by treatment with A, resulted in fourfold increase in luciferase activity compared with the untreated THP-1 cells. To correct for differences in transfection efficiency, THP-1 cells were cotransfected with the promoter less vector (pGL3 basic) and pRL-CMV vector containing the Renilla luciferase gene, as previously described by Mummidi and coworkers (23). These results indicated that A induces luciferase activity from CCR5 promoter in THP-1 cells. Because the pharmacological inhibitors PD-98059 (MEK-1/2) and SP-600125 (c-jun NH2- terminal kinase) attenuated A induced CCR5 mRNA expression, we examined whether THP-1 cells transfected with PA-3 luciferase construct elicited similar responses to A. As shown in Fig. 4B, PD-98059 reduced luciferase activity by 85%, whereas SP-600125 reduced activity by 45% in response to A. However, SB-203580 (a p38 MAP kinase inhibitor) that augmented A-induced CCR5 mRNA expression did not affect PA-3 luciferase activity in THP-1 transfected cells. Because PA-3 construct contains Egr-1 binding element of the CCR5 promoter but not the complete promoter region of CCR5, the effect of SB-203580 (p38 MAP kinase inhibitor) may occur through other promoter region of CCR5.
Chromatin immunoprecipitation assay demonstrates binding of Egr-1 to the CCR5 promoter.
To determine whether Egr-1 binds to the native chromatin in THP-1 monocytes, we performed chromatin immunoprecipitation assay (ChIP) assays on chromatin obtained from THP-1 cells, which were pretreated with A1–40 for 30 and 60 min. Chromatin samples were immunoprecipitated with antibody to either Egr-1 or SP-1. DNA recovered from the antibody-bound fractions and DNA from input chromatin (before immunoprecipitation) were analyzed by semiquantitative PCR-using primers (as shown in boxed region of Fig. 5A) corresponding to the promoter region of CCR-5 (from –847 to –603 relative to the transcription start site). As shown in Fig. 5B, THP-1 cells treated with A1–40 for 30 and 60 min exhibited increased amplification of PCR product, corresponding to the expected length (244 bp), with maximum Egr-1 chromatin binding activity at 30 min (Fig. 5B, lane 2). Both PD-98059 (Fig. 5B, lane 4) and SP-600125 (Fig. 5B, lane 5) reduced the in vivo Egr-1 chromatin binding activity in THP-1 cells treated with A1–40 by 75%. As a positive control, LPS increased Egr-1 binding to CCR5 promoter region (Fig. 5B, lane 6). Because the SP-1 binding element (–705 to –698) and Egr-1 binding element (–702 to –693) in CCR-5 promoter (–847 to –603) overlap, as indicated in Fig. 5A, we evaluated the SP-1 binding status to native chromatin derived from untreated and A-treated THP-1 cells. The data show (Fig. 5B, bottom) a PCR product corresponding to expected length (244 bp) in chromatin derived from untreated THP-1 cells, which were immunoprecipitated with antibody to SP-1. However, treatment with A modestly reduced amplification of PCR product (244 bp) at 30- and 60-min time periods (Fig. 5B, lanes 2 and 3, respectively). Both, PD-98059 (Fig. 5B, bottom, lane 4) and SP-600125 (Fig. 5B, bottom, lane 5) did not affect SP-1 chromatin binding activity in THP-1 cells treated with A1–40. Moreover, LPS also did not affect SP-1 binding to CCR5 promoter region (Fig. 5B, bottom, lane 6). Figure 5B, bottom, shows amplification of input DNA before immunoprecipitation. There is no change in the amplification of the input DNA in all the samples. Taken together, these data show the effect of A is specific for Egr-1 binding to CCR5 promoter region in vivo.
A-induced surface expression of CCR5 in THP-1 monocytes.
As shown in Fig. 6, A and B, A-treatment of THP-1 monocytes exhibited a time-dependent (1–4 h) increase in the surface expression of CCR5 as determined by flow cytometry. The optimal expression of CCR5 in response to A1–40 was 1.3- to 1.5-fold at 4 h (Fig. 6B). Similar increase in the surface expression of CCR5 was observed with A1–42 (data not shown). However, the surface expression of CCR2b remained unchanged in THP-1 cells, which were either untreated or treated with A (Fig. 6, C and D). It is pertinent to note that treatment of THP-1 cells with either A1–40 or A1–42 resulted in a time-dependent (4–24 h) increase in protein expression of both Egr-1 and CCR5 (Fig. 6E), whereas the protein levels of Erk-1 remained unchanged. The levels of Erk-1 in these samples show equal loading of protein in each lane.
Chemotactic response of THP-1 monocytes to -chemokines and A1–40.
Because interaction of A1–40 with THP-1 monocytes caused increased expression of CCR5, we determined whether -chemokines (MIP-1 and RANTES), known ligands of CCR5, could mediate chemotaxis of THP-1 monocytes. As shown in Fig. 7A, the presence of MIP-1 in the lower compartment of the Boyden chamber caused chemotaxis of THP-1 monocytes in a dose (10–40 ng/ml)-dependent manner. THP-1 monocytes that were pretreated with A1–40 (125 nM) for 4 h compared with untreated THP-1 cells showed an augmented (P < 0.05) response to chemotaxis in response to MIP-1 (40 ng/ml), consistent with the observed increased surface expression of CCR5 as shown in Fig. 6A. THP-1 monocytes also showed chemotaxis in response to RANTES as a chemoattractant, although A-treated THP-1 cells exhibited augmented chemotactic response to 10 and 20 ng/ml of RANTES (P < 0.05 and P < 0.001, respectively). It is pertinent to note that THP-1 monocytes underwent chemotaxis in response to A1–40, which was optimal at 125 nM concentrations (Fig. 7C). Furthermore, the chemotaxis of A-treated THP-1 monocytes was higher relative to untreated THP-1 monocytes in response to 125 nM of A as a chemoattractant (P < 0.05). It is pertinent to note that the chemotaxis of A-treated THP-1 monocytes was twofold higher with MIP-1 compared with chemotaxis with A1–40 (125 nM). We observed that application of MIP-1 (Fig. 7D), RANTES (Fig. 7E), and A1–40 (Fig. 7F) on both sides of the Boyden chamber nucleopore filter, at equivalent concentration of the chemoattractant, gave the same net results, as observed with background control. These results show that THP-1 cell migration in response to these chemoattractants was due to chemotaxis and not by chemokinesis.
Chemotaxis of THP-1 monocytes to MIP-1 is specific for CCR5 receptor.
Because A-induced the expression of CCR5 and CCR5 is a cognate receptor for MIP-1, we determined whether CCR5 expressed on monocytes played a role in chemotaxis. Thus we examined the effect of antibody to CCR5 on the chemotaxis of A-treated THP-1 monocytes in response to MIP-1 as a chemoattractant. As shown in Fig. 8, neutralizing antibody to CCR5 (5 µg/ml) reduced (>90%) chemotaxis of A-treated THP-1 monocytes (Fig. 8, lane 5); P < 0.001. However, antibody to CCR2b (5 µg/ml) (Fig. 8, lane 6) did not reduce MIP-1-mediated chemotaxis of A-treated THP-1 monocytes. As expected, the chemotaxis of A-treated THP-1 cells in response to MIP-1 was abrogated by antibody to MIP-1 (Fig. 8, lane 3; P < 0.001) but not with an antibody to MIP-1 (Fig. 8, lane 4).
Effect of transfection of THP-1 cells with Egr-1 siRNA on expression of CCR5.
Our recent studies (10) showed that both A1–40 and A1–42 activated transcription factors Egr-1 and AP-1 but not CREB or NF-B in THP-1 cells. Moreover, studies (10) showed that transfection of Egr-1 siRNA but not scEgr-1 siRNA in THP-1 cells were most effective at decreasing cellular Egr-1 protein level. Furthermore, we (10) showed that A1–40-induced Egr-1 DNA binding activity declined >65% in Egr-1 siRNA, but not in scEgr-1 siRNA, transfected THP-1 cells. As shown in Fig. 9, lane 3, A1–40-mediated increase in CCR5 mRNA expression in THP-1 cells was reduced by 90% in Egr-1siRNA-transfected THP-1 cells. However, in THP-1 cells transfected with scEgr-1 siRNA there was a modest decrease (<20%) in CCR5 transcripts compared with A1–40-treated THP-1 cells. As shown in Fig. 9, the CCR2a and CCR2b expression remained unaltered in THP-1 cells transfected with either Egr-1 siRNA or scEgr-1 siRNA. These results indicate that CCR5 expression but not CCR2a and CCR2b expression is presumably regulated by Egr-1 transcription factor.
A-induced Egr-1 DNA binding activity in THP-1 cells transfected with Egr-1 siRNA.
Because CCR5 mRNA expression was abrogated in THP-1 cells transfected with Egr-1 siRNA, in response to A1–40, we determined the Egr-1 DNA binding activity in nuclear extracts. As shown in Fig. 10, lane 1, there was absence of Egr-1-DNA complex when DNA probe was not added to the nuclear extract. However, A1–40 increased Egr-1 DNA binding activity by 5-fold in THP-1 cells (Fig. 10, lane 3, band 3) utilizing CCR5-Egr-1 (gcgggggtg) sequence compared with untreated cells (Fig. 10, lane 2). The presence of excess cold probe abolished these bands (1–3), indicating these bands are likely due to Egr-1 (Fig. 10, lane 6). The band corresponding to Egr-1 DNA complex (3) was super shifted in the presence of antibody to Egr-1 (Fig. 10, lane 7). Furthermore, the A-induced Egr-1 DNA binding activity was reduced in THP-1 cells transfected with Egr-1 siRNA (0.5 µg/ml) (Fig. 10, lane 4). This dose of Egr-1 siRNA has been previously shown by us to reduce (60%) Egr-1 protein levels in nuclear extracts of THP-1 cells (10). The specificity of Egr-1 siRNA in abrogating A-mediated Egr-1 DNA binding activity was validated by using scrambled Egr-1 siRNA (sc Egr-1 siRNA). The data in Fig. 10, lane 5, show that transfection of THP-1 cells with scEgr-1 siRNA does not have any inhibitory effect on A-induced Egr-1 DNA binding activity.
Effect of A on the chemotaxis of Egr-1 siRNA transfected THP-1 monocytes.
To determine the functional significance of the down regulation of CCR5 by Egr-1siRNA, we used these transfected cells for chemotaxis. As shown in Fig. 11, MIP-1 (20 ng/ml) exhibited an approximately fourfold increase in the chemotaxis of A1–40-treated THP-1 monocytes (Fig. 11, lane 2). The increase in the chemotaxis in A1–40-treated THP-1 monocytes could have occurred as a result of increased expression of CCR5. As shown in Fig. 11, lane 3, the chemotaxis of Egr-1 siRNA transfected THP-1 cells was reduced by 70% in response to MIP-1; P < 0.001. However, when THP-1 cells transfected with scrambled scEgr-1 siRNA were examined for chemotactic activity, there was no difference in migration compared with THP-1 cells treated with A (Fig. 11, lane 4).
|
DISCUSSION
|
|---|
In the present study, we demonstrate that amyloid peptides at submicromolar concentrations (125 nM), as found in the plasma of AD individuals (14), cause an increase in the gene expression of CCR5 in both PBM and human monocytic THP-1 cell line. Because amyloid peptide showed a similar profile of cytokine and chemokine gene expression (10) and CCR5 gene expression (present study) in both THP-1 cells and PBM, we used THP-1 monocytic cells as a model system of monocytes for studies described herein. We find that both nonfibrilar A1–40 and fibrilar A1–42 caused increased mRNA expression of CCR5. Moreover, we find that there is a time-dependent (1–4 h) increase in the surface expression of CCR5 but not of CCR2b in A1–40-treated THP-1 cells. It is pertinent to note that HIV-derived Tat protein has been shown to increase surface expression of CCR5 but not of CCR2 in peripheral blood mononuclear cells (31).
Next, we examined the A-mediated cell signaling mechanism leading to the increased CCR5 mRNA expression. We show that A1–40-induced expression of CCR5 mRNA is inhibited more than 90% by genistein (a protein tyrosine kinase inhibitor), PD-98059 and U-0126 (inhibitors of MEK1/2), and GW 5074 (a potent and specific inhibitor of c-Raf kinase). Furthermore, we observed that SP-600125, a specific inhibitor of c-Jun NH2 terminal kinase, reduced CCR5 mRNA expression by 60%. However, SB-203580 (a p38 MAP kinase inhibitor) augmented A-induced expression of CCR5 mRNA, although in transfection studies with CCR5-luc promoter SB-203580 did not alter CCR5 promoter activity. To address this discrepancy, further studies are required to delineate the role of p38MAP kinase, utilizing transfection with either dominant negative p38 MAP kinase (MEK3/MEK6) constructs or siRNA approach. In our previous study (10), we showed that A caused phosphorylation of tyrosine residues in a subset of proteins and phosphorylation of ERK-1/ERK-2 but not of p38 MAPK in THP-1. Taken together, these results suggest that A-induced cellular signaling for the expression of CCR5 involves activation of protein tyrosine kinase, c-Raf kinase, and MAPKK/MEK in THP-1 monocytes, as illustrated in Fig. 12A. However, the nature of putative receptor (e.g., RAGE, SR-A, CD36, CD47, and FRP-2) involved in the nonfibrillar and fibrillar form of A-mediated signaling in monocytes and microglia remains controversial (6, 18, 34, 35).
Subsequently, we analyzed the role of transcription factors, which may regulate the expression of CCR5 in THP-1 cells in response to amyloid peptide. The promoter region of CCR5 has been extensively studied (23, 24, 27), as CCR5 has been shown to be a co-receptor for the entry of macrophage-tropic strains of HIV-1. Studies have shown that the 5'-flanking region of CCR5 contains the cis-acting element sequences for NF-B, AP-1, SP-1, Oct-1, Oct-2, GATA-1, and CEBP transcription factors (22, 23, 27). However, we observed (10) that both soluble A1–40 and fibrilar A1–42 at submicromolar concentration caused increased activation of transcription factors, namely, early growth response-1 (Egr-1) and activator protein-1 (AP-1) but not of NF-B and CREB. It is pertinent to note that both A1–40 and A25–35 at micromolar concentrations (50–60 µM) have been shown to cause activation of NF-B in THP-1 monocytes (4). However, the amounts of circulating amyloid peptides in plasma of AD subjects have been observed to be in the submicromolar range (14), thus suggesting that studies should be undertaken at these physiological concentrations.
A computer-aided analysis of CCR5 promoter [the promoter sequence analysis reported by Mummidi et al. (23)] resulted in the identification of the cis-element GCGGGGGTG at positions –702 to –694, which closely resembles the bona fide Egr-1 binding sequence (GCGGGGGCG) with a change to T from C at position 8. It is pertinent to note that the macrophage colony-stimulating factor (M-CSF) gene promoter has a GCGGGGGAG sequence at position –273 to –265 that were found to be an Egr-1-binding site (29). Second, we show by EMSA analysis that there was a clear electrophoretic shift due to binding of Egr-1 protein to oligonucleotide probes corresponding to either the putative Egr-1 binding element (GCGGGGGTG) present in CCR5 promoter (present study) or to the bona fide Egr-1 oligonucleotide (GCGGGGGCG) in the previous study (10). A subsequent supershift analysis demonstrated that the A-induced transcription factor interacting with the oligonucleotide was Egr-1 but not SP-1. These results clearly demonstrate that nuclear Egr-1 interacts with the CCR5 promoter region.
A further proof that Egr-1 binds to the promoter region of the CCR5 gene was obtained by studying the effect of A in THP-1 cells transfected with truncated CCR5 promoter (–731 to +33) firefly luciferase construct. These studies indicated that the A responsive region of the CCR5 promoter is localized within the –731 to +33 regions, which contains a putative Egr-1 binding site and a SP-1 cis acting element. Finally, chromatin immunoprecipitation (ChIP) analysis demonstrated that Egr-1 binds in vivo to the CCR5 promoter and that interaction with this transcription factor increases after A treatment. Moreover, pharmacological inhibitors, which attenuated A-induced CCR5 expression, also reduced Egr-1 binding to CCR5 promoter in ChIP assay.
Finally, we observed that transfection of Egr-1 siRNA but not scrambled Egr-1 siRNA (scEgr-1 siRNA) in THP-1 cells caused >75% reduction in A1–40-mediated CCR5 expression. However, the mRNA expression of CCR2a and CCR2b was unaltered in THP-1 cells transfected with either Egr-1 siRNA or scEgr-1 siRNA. These results further corroborate our contention that CCR5 expression but not CCR2a and CCR2b expression is regulated by Egr-1 transcription factor. We previously (10) reported that A causes activation of ERK-1/2, which in turn results in phosphorylation of Elk-1. A parallel pathway involving activation of c-Jun NH2 terminal kinase also occurs, which causes phosphorylation of Elk-1 and AP-1 complex. Both of these pathways merge at Elk-1 phosphorylation resulting in the activation of Egr-1, as illustrated in Fig. 12A.
Because CCR5 is a cognate receptor for -chemokines (MIP-1 and RANTES), we hypothesized that this receptor could play a role in mediating chemotaxis of THP-1 monocytes. We show that A-activated monocytes, which exhibit increased surface expression of CCR5 but not of CCR2b, undergo chemotaxis in response to a chemoattractant gradient generated by chemokines (MIP-1 and RANTES) as well as to A1–40, although the extent of chemotaxis in response to amyloid (A1–40) was relatively lower compared with these chemokines. Here, we show that A-activated monocyte chemotaxis to MIP-1 is reduced in the presence of antibody to CCR5 but unaffected by antibody to CCR2b. Moreover, THP-1 monocytes transfected with Egr-1 siRNA, followed by treatment with A, which has been shown to reduce CCR5 expression, resulted in attenuated chemotaxis to MIP-1. These results indicate that chemotaxis of monocytes to MIP-1 requires cognate receptor CCR5 expressed on monocytes.
In conclusion, this is the first report, to our knowledge, showing that inhibition of Egr-1 transcription factor expression by Egr-1 siRNA can block A-mediated upregulation of CCR5 expression and concomitant chemotaxis of THP-1 monocytes. This is further supported by our finding of an Egr-1 like binding site in the promoter region of CCR5. We speculate that the increased presence of A peptides in plasma of AD patients (14) upregulates the surface expression of CCR5 on monocytes, which may facilitate their migratory response to the chemokines elaborated from activated microglia in brain parenchyma of AD. Both of these processes might be acting together (Fig. 12B) to promote the transmigration of monocytes across the BBB. Taken together, our studies show that downregulation of CCR5 gene expression by either Egr-1 siRNA or pharmacological agents (e.g., curcumin) (9), which reduce CCR5 expression, may provide a novel therapeutic approach to ameliorate the inflammation-induced progression of AD. It is pertinent to note that a 32-bp deletion in CCR5 gene (CCR5–32 mutant allele) results in a nonfunctional receptor, which has been shown to confer resistance to HIV infection and displays protective effect toward certain inflammatory diseases (3). However, a recent study by Combarros et al. (3), conducted in a small sample of AD patients in Spain, show that the CCR5–32 allele neither influences the risk for AD nor modifies the age at which the onset of disease occurs, indicating that the CCR5–32 allele is not a preventive factor for AD. Thus additional in vivo studies are warranted to determine the role of CCR5 in chemotaxis of monocytes across BBB in AD. Recently, several therapeutic strategies, such as immunization with amyloid peptides (16), which promote efflux of soluble A from brain to the periphery, have emerged, although the role of CCR5 in reducing amyloid burden in these studies is unknown. Our studies thus provide one avenue, among several approaches, to ameliorate amyloid peptide mediated inflammation and neurodegeneration in AD. This is supported by our finding that curcumin, a safe natural product, which suppresses CCR5 and cytochemokines expression in monocytes (9) in vitro has been shown to reduce levels of cytokines and plaque burden in Alzheimer's transgenic mice (17).
|
GRANTS
|
|---|
This work was supported by National Institutes of Health Grant POI-AG16233 and University of Southern California Grant ADRC-P50 AG 005142.
|
ACKNOWLEDGMENTS
|
|---|
We thank Dr. Stanley Tahara for critical reading of the manuscript. We appreciate Drs. Ralf Langen and Sajith Jayasinghe for help with analyzing the nonfibrillar and fibrillar forms of amyloid peptides using far-ultraviolet CD spectra and fluorescence spectroscopy. We thank Dr. Sunil Ahuja for kindly providing us with the CCR5 promoter construct used in this study.
|
FOOTNOTES
|
|---|
Address for reprint requests and other correspondence: V. K. Kalra, Dept. of Biochemistry & Molecular Biology, HMR-611, USC Keck School of Medicine, Los Angeles, CA 90033 (e-mail: vkalra{at}usc.edu)
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.
|
REFERENCES
|
|---|
1. Bamshad MJ, Mummidi S, Gonzalez E, Ahuja SS, Dunn DM, Watkins WS, Wooding S, Stone AC, Jorde LB, Weiss RB, and Ahuja SK. A strong signature of balancing selection in the 5' cis-regulatory region of CCR5. Proc Natl Acad Sci USA 99: 10539–10544, 2002.[Abstract/Free Full Text]
2. Bennett BL, Sasaki DT, Murray BW, O'Leary EC, Sakata ST, Xu W, Leisten JC, Motiwala A, Pierce S, Satoh Y, Bhagwat SS, Manning AM, and Anderson DW. SP600125, an anthrapyrazolone inhibitor of Jun N-terminal kinase. Proc Natl Acad Sci USA 98: 13681–13686, 2001.[Abstract/Free Full Text]
3. Combarros O, Infante J, Llorca J, Pena N, Fernandez-Viadero C, and Berciano J. The chemokine receptor CCR5-
32 gene mutation is not protective against Alzheimer's disease. Neurosci Lett 366: 312–314, 2004.[CrossRef][ISI][Medline]
4. Combs CK, Karlo JC, Kao SC, and Landreth GE. -Amyloid stimulation of microglia and monocytes results in TNF-dependent expression of inducible nitric oxide synthase and neuronal apoptosis. J Neurosci 21: 1179–1188, 2001.[Abstract/Free Full Text]
5. Eglitis MA and Mezey E. Hematopoietic cells differentiate into both microglia and macroglia in the brains of adult mice. Proc Natl Acad Sci USA 94: 4080–4085, 1997.[Abstract/Free Full Text]
6. El Khoury J, Hickman SE, Thomas CA, Cao L, Silverstein SC, and Loike JD. Scavenger receptor-mediated adhesion of microglia to beta-amyloid fibrils. Nature 382: 716–719, 1996.[CrossRef][Medline]
7. Franchin G, Zybarth G, Dai WW, Dubrovsky L, Reiling N, Schmid
TOP
ABSTRACT
METHODS
). B: one of the potential pathways (modified from Ref.