|
|
||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Department of Microbiology, Sapporo Medical University School of Medicine, Sapporo 060-8556, Japan
1Correspondence: Department of Microbiology, Sapporo Medical University School of Medicine, South-1, West-17, Chuo-ku, Sapporo 060-8556, Japan. E-mail: fujii{at}sapmed.ac.jp
| ABSTRACT |
|---|
|
|
|---|
B and AP-1 was suppressed in monocytes infected with measles virus, but not in infected epithelial cells. This cell-type-specific suppression of the inflammatory response represents a potential for measles virus to evade host immune system. In the current study, we examined the suppression mechanism of lipopolysaccharide (LPS)-induced, namely Toll-like receptor 4 (TLR4)-mediated, activation of NF-
B and AP-1 in measles virus-infected monocytic cells. In the infected cells, LPS treatment failed to induce the formation of active protein kinase complex containing TAK1, TAB2 and tumor necrosis factor receptor-associated factor 6 (TRAF6), dissociate from TLR complexes containing Interleukin-1 receptor-associated kinase 1 (IRAK1). Ubiquitin-modifying enzyme A20, which is a host negative feedback regulator of NF-
B, was dramatically up-regulated in infected monocytic cells, but not in infected epithelial cells. Suppression of A20 expression by siRNA restored LPS-induced signaling in infected cells. Measles virus phosphoprotein (P protein) expression was necessary and sufficient for the induction of A20. P protein interacted indirectly with a negative regulatory motif in the A20 gene promoter, and released the suppression of A20 transcription, independent of the activation of NF-
B.—Yokota, S., Okabayashi, T., Yokosawa, N., Fujii, N. Measles virus P protein suppresses Toll-like receptor signal through up-regulation of ubiquitin-modifying enzyme A20.
Key Words: viral infection immunosuppression TLR monocyte NF-
B
| INTRODUCTION |
|---|
|
|
|---|
B (NF-
B) and activator protein-1 (AP-1) was dramatically suppressed in MeV-infected monocytic cells but not in infected epithelial cells (11)
In this study, we examined a putative molecular mechanism of suppression of LPS-induced NF-
B and AP-1 activation by MeV in monocytic cells. LPS interacts with a receptor complex containing Toll-like receptor (TLR) 4, MD-2 and CD14 on the surface of mammalian cells, which transduces intracellular signals that result in the activation of NF-
B and production of IFN (12)
. We examined the effect of MeV on the TLR-mediated signal transduction in monocytic cells.
| MATERIALS AND METHODS |
|---|
|
|
|---|
Semiquantitative reverse transcription-polymerase chain reaction (RT-PCR)
Total RNA was isolated from cells using an RNeasy Mini kit (Qiagen, Hilden, Germany). The OneStep RT-PCR kit (Qiagen) were used for RT-PCR. Various amounts of RNA were used to create a standard curve to confirm that quantitation was performed. The primer set for detecting A20 were as follows: forward, 5'-CATCAGTGCCACTTCTCAGT-3' and reverse, 5'-GCTGCTATAGCCGAGAACAA-3'. The primer sets for detecting all other targets were previously described (10)
.
Expression plasmids encoding MeV proteins
We obtained a cDNA of the MeV P protein by RT-PCR, using RNA derived from U937 cells infected with MeV Hälle. RT-PCR was carried out with forward primer MeVP-F: 5'-GAGCCGATGGCAGAAGAGCA-3' and reverse primer MeVP-R: 5'-GTTGAGCTGTAGCTACTTC-3'. The amplified PCR product was inserted into pTARGET plasmid (Promega, Madison, WI), subcloned, and sequenced. From nucleotide sequence data, cDNAs that did not have a G inserted at nucleotide 751 were used to generate expression plasmids for P and C proteins. The cDNA containing a G insertion at nucleotide 751 was used to generate the expression plasmid for V protein. To disrupt the C protein initiation codon (to generate the expression plasmids for P and V proteins), we carried out the PCR using forward primer: 5'-GAGCCGATGGCAGAAGAGCAGGCACGCCGCGTCAAAAACGG-3' and the reverse primer MeVP-R. To disrupt initiation codon of P and V proteins (to generate the expression plasmid for C protein), we carried out the PCR using forward primer: 5'-GAGCCGGCGGCAGAAGAGCAGGCACGCCATGTCAAAAACGG-3' and the reverse primer MeVP-R. PCR products were inserted into pTARGET, and then subcloned. Protein expression driven by each plasmid was confirmed using an in vitro translation system (TNT T7-coupled reticulocyte lysate system, Promega, Madison, WI, USA).
Plasmids were transfected into cultured cells using the Superfect reagent (Qiagen), according to the manufacturers instructions.
TLR agonist treatment
THP-1 and U937 cells were pretreated with 1 x 10–8 M 1
, 25-dihydroxyvitamin D3 (active vitamin D3) for 16 h (unless otherwise noted) before LPS treatment. Suspension cells were inoculated at 5 x 105 cells /ml. Adherent cells were grown to >90% confluence. Cells were treated with 100 ng/ml LPS (unless otherwise noted) derived from Escherichia coli O111:B4 (Sigma-Aldrich, St. Louis, MO, USA).
Expression of small interfering (siRNA) targeting A20
Expression plasmid encoding A20-specific siRNA and or Renilla luciferase-specific siRNA (as a control) were kindly provided from Dr. Shoji Yamaoka (Tokyo Medical and Dental University, Tokyo, Japan) (16)
. Cells were transfected using the Superfect reagent (Qiagen), according to the manufacturers instructions, cultured for 48 h in the presence of active vitamin D3, and then treated with LPS as above.
Western blot
The mouse monoclonal antibody against A20 (clone 59A426) was purchased from Stressgen (Victoria, BC, Canada). Rabbit antibodies against TAK1-binding protein 2 (TAB2; H-300), tumor necrosis factor receptor-associated factor 6 (TRAF6; H-274) and A20 (H-100), and mouse monoclonal antibodies against interleukin-1 receptor-associated kinase 1 (IRAK1; F-4) were purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA). Rabbit anti–transforming growth factor-β activated kinase 1 (TAK1) antibody was purchased from Upstate Biotechnologies (Lake Placid, NY, USA). Mouse antimulti-ubiquitin monoclonal antibody (clone FK2) was purchased from MBL (Nagoya, Japan). Mouse monoclonal antibody against MeV P protein (clone 49–21) was purchased from Argene (Varilhes, France). Rabbit polyclonal antibodies against MeV V and C proteins (9)
were kindly provided by Dr. Kaoru Takeuchi (Tsukuba University, Tsukuba, Japan). Guinia pig polyclonal anti-MeV antibody was purchased from Denka Seiken (Tokyo, Japan).
Preparation of total cell lysates, SDS-polyacrylamide gel electrophoresis (SDS-PAGE) and Western blotting were carried out as described previously (10)
. Alkaline phosphatase-conjugated goat anti-rabbit or anti-mouse immunoglobulin antibodies (BioSource International, Camarillo, CA, USA) and alkaline phosphatase-conjugated F(ab')2 fragment of goat antiguinea pig IgG(H+L) (Jackson ImmunoResearch, West Grove, PA, USA) were used as secondary antibodies. Bromochloroindolylphosphate/nitro blue tetrazolium was used as an enzyme substrate.
Immunoprecipitation
Cells were lysed with RIPA buffer (0.5% Triton X-100, 0.1% sodium deoxycholate, 100 mM NaCl, 4 mM EDTA, 1 mM phenylmethansulfonyl fluoride, 0.2 mU/ml aprotinin, and 50 mM Hepes-NaOH pH 7.5). After preabsorption of with protein G-Sepharose (GE Healthcare UK, Buckinghamshire, England), an aliquot (50 µl) of the lysate was mixed with an antibody [5 µl of mouse monoclonal antibody against TRAF6 (D-10) or goat polyclonal antibody against TAB2 (K-20) (Santa Cruz Biotechnology)] and then incubated at 4°C for 1 h. Protein G-Sepharose was added to the mixture, incubated at 4°C overnight with constant rotation, and washed repeatedly in RIPA buffer. In the case of precipitating only the target protein, namely disassembling interacting proteins, RIPA buffer adding 0.05% SDS was used as wash buffer (17)
. Proteins bound to the resin were extracted by boiling in SDS-PAGE sample buffer and separated by SDS-PAGE. The targeted proteins were detected by Western blotting described above.
Analysis of proteins that interacted with A20 promoter ELIE-motif containing oligonucleotides
The 5'-biotin-labeled double-stranded oligoDNA containing ELIE motif (5'-AGTCACGTGACTTTGGAAAG-3') and its mutant (m3: 5'-AGTCACACGAGGACGGAAAG-3') were chemically synthesized. The cell lysate of MeV-infected THP-1 cells were prepared according to the method described for immunoprecipitation. Biotin-labeled oligonucleotide (2 µg) was added to the cell lysate, and incubated at 4°C for 1 h on a rotator. Twenty microliters of a suspension of ImmunoPure Immobilized Streptavdin Gel (Pierce, Rockford, IL, USA) were added and the mixture was incubated at 4°C for 1 h with a constant rotation. The resin was washed repeatedly in RIPA buffer. Proteins bound to the resin were extracted by boiling in SDS-PAGE sample buffer, and separated by SDS-PAGE. The MeV proteins were detected by Western blot using a guinia pig polyclonal anti-MeV antibody.
Enzyme-linked immunosorbent assay (ELISA)
IL-8 in culture supernatants was quantified using a Human CXCL8/IL-8 DuoSet ELISA Development kit (R&D systems, Minneapolis, MN, USA).
Luciferase reporter gene assay
Expression mechanism of A20 was examined using a dual luciferase reporter assay. The reporter plasmids harboring A20 wild promoter and its mutant derivatives just upstream of firefly luciferase gene (18)
were kindly provided by Dr. Rivka Dikstein (The Weizmann Institute of Science, Rehovot, Israel). THP-1 cells or THP-1 cells persistently infected with MeV were cotransfected with reporter plasmid (1 µg) and a reference plasmid (0.1 µg of pRL-TK, harboring the HSV thymidine kinase promoter just upstream of Renilla luciferase, Promega) using the Superfect reagent (Qiagen) according to the manufacturers instructions. After cultivation for 24 h, the cells were treated with active vitamin D3 for 12 h and then incubated in the presence or absence of 100 ng/ml LPS for 12 h. The cells were lysed, and firefly luciferase and Renilla luciferase activities were measured. Experiments were performed in triplicate, and reporter activity is present as the ratio of firefly luciferase activity to Renilla luciferase activity.
| RESULTS |
|---|
|
|
|---|
|
In immunoblot analysis, electrophoretic mobility of TAK1, TAB2, and IRAK1 were shifted slower in response to LPS treatment, probably because these molecules was phosphorylated. Such a shift was not observed in MeV-infected cells even after LPS treatment. (Fig. 1)
.
In our previous report (11)
, no significant and common changes of expression levels of TLR4, CD14 and MD2 were observed between MeV-infected and uninfected monocytic cell lines (U937 and THP-1). These suggest that the impairment of LPS-induced responses is not due to down-regulation of receptor components for LPS.
Transcriptional activation of the NF-
B negative regulator A20 by MeV
The suppression of NF-
B and AP-1 activation by MeV infection seems to be located upstream of the formation of active TAK1 complexes, which is a common signaling pathway of the activation of both transcription factor. We performed a screen for the known host-cell negative regulators of the TLR signaling pathway (reviewed in ref. 19
). We determined the expression levels of short form of MyD88, IRAK-M, SOCS1, Tollip, ST2L, SIGIRR, TRIAD3A and A20 in MeV-infected and uninfected monocytic cell lines by RT-PCR. Except A20, significant differences in expression levels of these NF-
B negative regulators were not observed between infected and uninfected cells (data not shown). The ubiquitin-modifying enzyme A20 was up-regulated in monocytic cell lines (U937 and THP-1) but not in epithelial cell lines (SiHa and OSC70), during persistent MeV infection (Fig. 2
A). Up-regulation of A20 in monocytic cells was also observed during acute infection (Fig. 2B
), during which there was a continuous increase in the protein and mRNA levels of A20. A20 protein levels were nearly undetectable in SiHa and OSC70 cells during acute infection, whereas A20 mRNA levels transiently increased and reached a peak of expression 6 h postinfection. Previously (11)
, we showed that suppression of NF-
B activation occurs following MeV infection but not mumps virus infection. U937 cells persistently infected with mumps virus high levels IL-8 (11)
, and did not exhibit induction of A20 (data not shown).
|
To further characterize the role of MeV-induced A20 expression in monocytic cells, we transfected THP-1 cells with an expression plasmid that encoded an A20-specific siRNA. Expression of the A20-specific siRNA but not a control siRNA (specific for luciferase) reduced the levels of basal expression in MeV-infected cells and induction of A20 in response to LPS in infected and uninfected cells (Fig. 3
A). In cells expressing A20 siRNA, LPS-induced IL-8 production (Fig. 3B
) in MeV-infected cells was similar to uninfected cells, whereas the control siRNA had no effect on LPS-induced IL-8 production.
|
A20 is a ubiquitin-modifying enzyme. One of the targets for A20 is TRAF6 (20)
. We examined the ubiquitination state of TRAF6 in MeV-infected cells (Fig. 4
A). TRAF6 was immunoprecipitated under high-stringency conditions to dissociate any TRAF6-interacting proteins and then examined by Western blot. Polyubiquitinated TRAF6 was present in uninfected cells after LPS stimulation but not in infected cells. Furthermore, suppression of the LPS-induced polyubiquitination of TRAF6 in MeV-infected cells was restored by the expression of A20 siRNA (Fig. 4B
). These results indicated that MeV up-regulates A20 expression, which results in defective ubiquitination and recruitment of TRAF6.
|
Identification of a promoter element that contributes to MeV-induced transcriptional activation of A20
To determine the mechanism of A20 up-regulation, we carried out a reporter gene assay of A20 promoter activity in MeV-infected and uninfected THP-1 cells (Fig. 5
A, B). The A20 promoter has been characterized in detail by Ainbinder et al. (18)
. They identified two functional NF-
B-binding motifs and a negative regulatory element termed ELIE, which is located upstream the NF-
B-binding motifs. Basal (unstimulated) transcriptional activity of the wild-type A20 promoter was markedly higher in MeV-infected monocytic cells compared to uninfected cells (Fig. 5B
). This high level of transcriptional activity in infected cells was decreased by the introduction of mutations in the negative regulatory ELIE motif (mutants m1 to m4). The latter nucleotide residues of the ELIE motif seemed to be important for the transcriptional activation by MeV, because m2, m3, and m4 mutant showed more highly decreased promoter activity than m1 mutant. Mutation of the two NF-
B motifs did not have as much of an effect on promoter activity in MeV-infected cells. However, mutations in NF-
B binding motifs seem to affect on the function of ELIE. Because mutations in NF-
B motifs partly decreased A20 promoter activity and promoter with double mutations in both NF-kB and ELIE (m1 mutant) motifs showed slightly higher promoter activity than promoter with the mutation in ELIE (m1 mutant). These results suggested that the negative regulatory ELIE motif is important for MeV-induced A20 up-regulation.
|
MeV P protein interacts with the ELIE motif in the A20 promoter and up-regulates A20 expression
To identify proteins that could interact with the ELIE motif of the A20 promoter, we synthesized biotin-labeled double-stranded oligonucleotides containing the wild-type ELIE motif and a mutated form of ELIE, ELIEm3. Biotinylated oligonucleotides were mixed with whole cell lysates derived from MeV-infected THP-1 cells, and then precipitated with streptavidin-conjugated beads. Protein-oligonucleotide complexes were analyzed by Western blot using an anti-MeV polyclonal antiserum (Fig. 5C
). A protein of
60 kDa specifically interacted with the wild-type ELIE motif of the A20 promoter. The ELIE-interacting protein had an electrophoretic mobility that was closely similar to native MeV P protein synthesized by in vitro translation (Fig. 5C
) and was recognized by an anti-P protein monoclonal antibody (data not shown). MeV N protein precipitated nonspecifically with both wild-type ELIE and ELIEm3. We were unable to detect an interaction between ELIE oligonucleotides and P protein synthesized by in vitro translation (data not shown). These results indicated that MeV P protein functionally interacts with the ELIE motif of the A20 promoter and that the interaction is indirect.
To confirm the functional relationship between the ELIE and MeV P protein, we examined the activity of the ELIE motif in a heterologous context (Fig. 6
). Reporter plasmids in which small regions of the A20 promoter, either the complete ELIE motif (nucleotides –74 to –34 relative to the transcriptional start site) or ELIE with a 3 nucleotide deletion (–71 to –34) are inserted upstream of an
-actin core promoter (nucleotides –38 to +80) (ELIE-actin and mELIE-actin, respectively) that drives a luciferase reporter gene (18)
. THP-1 cells were cotransfected with a reporter plasmid, and a expression plasmid encoding P, V, or C protein. Luciferase activity driven by the ELIE-actin promoter was lower than that driven by the mELIE-actin promoter. Coexpression of P protein, but not V or C protein, specifically up-regulated to transcriptional activity of the ELIE-actin promoter, and the level of up-regulated activity was comparable to mELIE-actin promoter activity in mock-transfected cells (Fig. 6B
). Coexpression of any viral proteins did not exhibit effect on mELIE-actin promoter activity. These results indicated that the ELIE motif of the A20 promoter can suppress the activity of a heterologous promoter, and that MeV P protein releases the negative regulatory effect of the ELIE motif.
|
Contribution of MeV P protein to the suppression of the inflammatory response in monocytic cells
To examined the effect of MeV P protein on the immune response of monocytic cells, THP-1 cells were transfected with expression plasmids encoding P, V, or C protein, which are expressed from the same genome region of P protein. Expression of P protein resulted in up-regulation of A20 (Fig. 7
B), and suppression of LPS-induced formation of complex containing TAB2 and TRAF6 (Fig. 7C
), polyubiquitination of TRAF6 (Fig. 7D
), and IL-8 production (Fig. 7E
), similar to effect of MeV infection. IL-8 production was up-regulated in cells expressing V protein in the presence and absence of LPS stimulation. These results indicated that P protein suppresses NF-
B activation via up-regulation of A20, whereas expression of V protein has somewhat opposite effect to P protein. V protein activated NF-
B-mediated signals, including up-regulation of IL-8 and A20, whereas P protein up-regulated A20 independent to activated NF-
B. C protein had no effect on A20 expression or the induction of proinflammatory cytokine induction.
|
Suppression of LPS-induced IL-8 in the P-protein-expressing cells restored by the expression of A20 siRNA (Fig. 8
). The expression levels of P protein in monocytic and epithelial cells during MeV infection were also examined by Western blot (Fig. 2B
). Similar to the level of expression of A20, expression of P protein in monocytic cells was higher than in epithelial cells. This result provided additional evidence of a positive functional link between P protein expression and A20 induction.
|
| DISCUSSION |
|---|
|
|
|---|
B and AP-1 activation was suppressed in MeV-infected monocytic cells but not in infected epithelial cells (11)
B, A20. siRNA suppression of A20 in MeV-infected cells indicated that A20 induction is both necessary and sufficient for NF-
B and AP-1 suppression in MeV-infected cells. A20 is an NF-
B target gene and contributes to a negative feedback mechanism of regulation of NF-
B, which is activated by TLR (20
(22)
B activation by modifying the ubiquitination state of the A20 target proteins, which contribute to signal transduction of NF-
B activation. A20 has two ubiquitin-editing domains, an N-terminal ovarian tumor (OUT) domain that mediates deubiquitination and a C-terminal zinc finger motif that act as a ubiquitin ligase (26)
Our results revealed a novel mechanism of regulation of NF-
B in MeV-infected cells, which involves up-regulation of A20 by MeV. Transcription of A20 is typically activated by NF-
B as part of a negative feedback mechanism of regulation. However, in MeV-infected monocytic cells, NF-
B is not activated. Ainbinder et al. (18)
hypothesized that a factor that binds to the ELIE motif in the A20 promoter inhibits basal transcription during elongation, via DRB sensitivity-inducing factor, and that activated NF-
B rapidly releases this repression. Based on our results, we propose that MeV P protein indirectly interacts with the ELIE motif of A20, and releases the repression of A20 transcription instead of activated NF-
B. P protein has been shown to have multiple functions. It binds to MeV L (RNA polymerase), N proteins, and RNA to form the replicase complex. The N-terminal portion of N protein and the N-terminal portion of L protein interact with the C-terminal portion of P protein (28
29
30)
. P protein functions as a molecular chaperone, namely stabilizes L protein (30)
. It sequesters N protein in the cytoplasm (31)
and regulates nucleocapsid assembly (32)
. There is also evidence that P protein interacts with host-cell proteins. Chen et al. (33
, 34)
reported that P protein interact with human p53-induced-RING-H2 (PIRH2), which is a ubiquitin E3 ligase, and specifically stabilizes PIRH2 in vivo by preventing ubiquitination of the protein. Our results indicate that P protein activates transcription of A20 by releasing suppression mediated by the ELIE motif of the A20 promoter. P protein interacted with the ELIE motif, however, the interaction appeared to be indirect, most likely via an ELIE-binding protein. Chen et al. (29)
also showed that P protein functions in transcriptional activation using a reporter gene assay in a yeast two hybrid system. In this previous study, the yeast Gal4 DNA binding domain-conjugated to full-length P or the N-terminal portion of P protein activated reporter gene expression in the absence of the Gal4 activation domain-containing proteins. These results suggested that the N-terminal portion of P protein plays a role in the activation of transcription.
P protein is encoded by the P/V/C gene of the MeV genome. The mRNA from this gene encodes three different proteins: P, V, and C. V protein shares an initiation codon and 231 N-terminal amino acid residues with the P protein. RNA editing inserts a gusnosine (G) at position 751 of the P/V/C mRNA (corresponding to position 2499 of the MeV genome) results in the generation of a unique C-terminal region in V protein (35)
. C protein is generated from the same mRNA but is translated from a different initiation codon (36)
; thus, the amino acid sequence of C protein is completely different from P and V proteins. Whereas V protein did not enhance A20 transcriptional activity, expression of V protein by transient transfection resulted in activation of NF-
B and enhanced production of IL-8. Thus, V protein, which has the same N-terminal amino acid sequence as P protein, appears to have the opposite effect of P protein on the inflammatory response. Tober et al. (37)
reported that V protein efficiently competes with P protein for binding to MeV N protein. The authors suggested that V protein acts to balance the accumulation of viral gene products in cell culture, depending on whether it interacts with MeV N protein. Our results should be additional evidence of the competitive action of P and V proteins.
In conclusion, we showed that MeV P protein activates the transcription of a host-cell negative regulator of NF-
B, A20, in an NF-
B-independent manner (Fig. 9
). Up-regulation of A20 was observed in monocytic cells, but not in epithelial cells. This cell-type specificity is most likely due to difference in the expression levels of P protein and/or the balance of P and V proteins in the cells. Suppression of the inflammatory response, namely NF-
B and AP-1 activation, results in immunological silencing. Silent monocytes carrying MeV may escape host defense response and serve as mechanism for dissemination of the virus throughout the body via blood and lymphatic fluids
|
| ACKNOWLEDGMENTS |
|---|
Received for publication May 8, 2007. Accepted for publication July 19, 2007.
| REFERENCES |
|---|
|
|
|---|
RII and inhibits human B cell antibody production. J. Exp. Med. 186,269-278
/β but not IFN-
signaling by inhibiting STAT1 and STAT2 phosphorylation. FEBS Lett. 545,177-182[CrossRef][Medline]
signaling pathway: suppression of Jak1 phosphorylation and association of viral accessory proteins, C and V, with interferon-
receptor complex. Virology 306,135-146[CrossRef][Medline]
B and AP-1 activation in monocytic cells persistently infected with measles virus. Virology 361,294-303[CrossRef][Medline]
down-regulates Hsp27 expression and suppresses the negative regulation of cell death in oral squamous cell carcinoma lines. Cell Death Differ. 10,313-322[CrossRef][Medline]
B. Mol. Cell. Biol. 24,2444-2454
B signalling. Nature 430,694-699[CrossRef][Medline]
B activation in airway epithelium upon influenza virus infection. Eur. J. Pharmacol. 541,198-204[CrossRef][Medline]
B activation by dual ubiquitin-editing functions. Trends Biochem. Sci. 30,1-4[CrossRef][Medline]This article has been cited by other articles:
![]() |
M. Carsillo, K. Klapproth, and S. Niewiesk Cytokine Imbalance after Measles Virus Infection Has No Correlation with Immune Suppression J. Virol., July 15, 2009; 83(14): 7244 - 7251. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |