| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Regular Articles |
From the Department of Pathology,*
Yale University
School of Medicine, New Haven, Connecticut; Department of
Pathology,
European Institute of Oncology and
University of Milan School of Medicine, Milan, Italy; and Cancer
Research Center,
The Burnham Institute, La
Jolla, California
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
and a ß
subunit.5-7
In addition to mediating physical
interactions, integrins signal intracellularly through their
cytoplasmic domains.8
The cytoplasmic domains also control
integrin affinity for ligands and subcellular
localization.8-10
The cytoplasmic domain of the ß1 subunit, in its
canonical form (ß1A), is highly conserved from fungi and
invertebrates to vertebrates.11
Experimental modifications
of the ß1 cytoplasmic domain have been shown to affect
cell proliferation,12,13
development,14
migration,15,16
integrin localization,17,18
and
mitogen-activated protein kinase activation19,20
and
phosphorylation of focal adhesion kinase or
paxillin.15,21-23
Alternative splicing events, by creating
variant cytodomains, generate functionally distinct integrin
complexes.8
Alternatively spliced forms of the ß
(ß1, ß3, ß4, and
ß5) and (3, 6, and
7) integrin cytoplasmic domains have been
identified,8,24
thus adding further complexity to the
regulatory pathways mediated by integrins.
The integrin ß1 variants are known as: the canonical ß1A, ß1B, ß1C, ß1C-2, and ß1D; they differ in the COOH-terminal sequences.8 The ß1C integrin (formerly ß1S)25 is generated by the presence of an unspliced intervening 116-bp sequence (exon C), which causes a frame shift in the 3' end of the ß1 subunit and codes for a unique 48-amino acid COOH-terminal sequence. Unlike ß1A, ß1C inhibits cell proliferation and is downregulated in cancer cells.12,13,26
We show here that ß1C is expressed in a subset of epithelial cells exhibiting a nonproliferating and differentiated phenotype, and that it is lost in regenerative areas of prostate glands. We also provide evidence suggestive of a causal role for the ß1C cytodomain in suppressing epithelial cell proliferation.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The PC3 human prostatic carcinoma cell line was obtained from American Type Culture Collection (Manassas, VA). PC3 cells were maintained in RPMI (Life Technologies Inc., Gaithersburg, MD) supplemented with 10% heat-inactivated fetal calf serum (FCS) (Gemini Bioproducts Inc., Calabasas, CA), 2 mmol/L glutamine (Gemini Bioproducts Inc.), 100 µg/ml streptomycin-100 U/ml penicillin (Gemini Bioproducts Inc.), 0.1 mmol/L nonessential amino acids (Life Technologies), and 1 mmol/L sodium pyruvate (Life Technologies). Human prostate luminal epithelial cells, BPH-1,27 were maintained in RPMI supplemented with 2.5% FCS, 2 mmol/L glutamine, and 100 µg/ml streptomycin-100 U/ml penicillin.
Rabbit antibodies specific for the ß1C or the ß1A subunit cytoplasmic domains were generated and affinity purified as previously described.26 The following antibodies were used: fluorescein isothiocyanate-conjugated rat monoclonal antibody to mouse CD4, clone YTS 191.1 (Caltag Laboratories, San Francisco, CA); mouse monoclonal antibody to the human ß1 integrin extracellular domain, mAb 13 (Beckton Dickinson, San Jose, CA) and kindly provided by Dr. K. M. Yamada (National Institutes of Health, Bethesda, MD); mouse monoclonal antibodies to basal cytokeratins 1, 5, 10, and 14, clone 34ßE12 (Enzo Diagnostic, Farmingdale, NY) or to type I and type II cytokeratins, clone AE1/AE3 (Boehringer Mannheim, Indianapolis, IN); and mouse monoclonal antibody to cytokeratins 8 and 18, clone CAM 5.2 (Becton Dickinson). Fluorescein isothiocyanate-rat immunoglobulin G (IgG) were purchased from Caltag.
Tissue Specimens
Samples (autopsy or surgical specimens) were obtained from the files of the Department of Pathology at Yale New Haven Hospital. The following adult tissues were included in the study: benign prostate (62 samples, including 25 benign prostatic hyperplastic samples), normal liver (12 samples), normal gallbladder (4 samples), normal kidney (5 samples), and normal lung (1 sample). Developing tissues from spontaneous or therapeutic abortions corresponding to different gestational ages were also included: prostate (4 samples, 18 to 22 weeks; 9 samples, 23 weeks or later) and liver (3 samples, 16 to 19 weeks). Hematoxylin and eosin sections were analyzed from all samples to assess the integrity of the tissue selected for the evaluation of ß1C, ß1A, or cytokeratin immunoreactivity. All tissues were obtained under review board-approved protocols.
Immunohistochemistry
Immunostaining of ß1C, ß1A, and cytokeratins was performed essentially as described,26 with minor modifications. For immunostaining of liver specimens, nonspecific binding of biotin/avidin was prevented using the Avidin/Biotin blocking kit from Vector Laboratories (Burlingame, CA), according to the manufacturer's instructions. For immunostaining of prostate specimens, endogenous peroxidase was quenched with 3% H2O2 for 5 minutes at room temperature, microwave treatment was avoided, blocking was achieved with either 50% goat or horse serum in Tris-buffered saline containing 0.2% bovine serum albumin for 20 minutes at room temperature; each incubation step was followed by three washes with Tris-buffered saline. The specificity of the affinity-purified antibody to the ß1C 785 to 808 peptide12 in immunohistochemical staining was confirmed as follows: before the immunostaining procedure, the affinity-purified antibody to ß1C was preincubated for 30 minutes at 4°C with 10 to 30 µg/ml of either ß1C 785 to 808 peptide or a control fibrinogen-derived peptide.28
Immunoblotting and Immunoprecipitation
Frozen prostate tissue, obtained from autopsy specimens, was homogenized using lysis buffer containing 100 mmol/L Tris, pH 7.5, 150 mmol/L NaCl, 0.1% Triton X-100 (Sigma Chemical Co., St. Louis, MO), 5% sodium dodecyl sulfate (SDS; American Bioanalytical, Natick, MA), 1 mmol/L phenylmethylsulfonyl fluoride (Life Technologies), 10 µg/ml leupeptin (Calbiochem, San Diego, CA), 1 mmol/L benzamidine (Sigma), 1 µmol/L D-phenylalanyl-L-prolyl-L-arginine chloromethyl ketone (Boehringer Mannheim), 10 µg/ml soybean trypsin inhibitor (Life Technologies), using an OMNI 2000 homogenizer (OMNI International Inc., Gainesville, VA). Insoluble material was removed by centrifugation at 14,000 x g for 30 minutes at 4°C. The protein content in each lysate was quantitated using the BCA protein assay reagent (Pierce, Rockford, IL) according to the manufacturer's instructions. To detect ß1C integrin, 200 µg of detergent tissue extract was electrophoresed on a 7.5% SDS-polyacrylamide gel under reducing conditions, transferred to nitrocellulose (Schleicher & Schuell, Keene, NH), and immunostained as described,26 using either 5 µg/ml affinity-purified antibody to ß1C; 5 µg/ml nonimmune rabbit IgG; or 2.5 µg/ml mAb 13, monoclonal antibody to ß1 integrin.
BPH-1 cells were surface iodinated and proteins immunoprecipitated using rabbit antisera to ß1C or to ß1A or using normal rabbit serum, as previously described.12 The immunocomplexes were separated on a 7.5% SDS-polyacrylamide gel under reducing conditions and visualized by autoradiography.
Inducible Expression of CD4-ß1C and CD4-ß1A Chimeric Constructs
HindIII DNA fragment, encoding the cytoplasmic domain of ß1C (nucleotides 2357 to 2613),25 was isolated from the pBJ1-ß1C plasmid. The fragment was inserted into the HindIII site of the Ch2 chimera described by Lukashev et al22 which consists of the extracellular domain of murine CD4 joined to the transmembrane domain of the ß1 integrin. The resulting construct was designated Chß1C. Correct assembly of the construct was verified by nucleotide sequencing. The Ch1 chimeric construct, designated here Ch1ß1A, containing the extracellular domain of murine CD4 and the transmembrane and cytoplasmic domains of the ß1A integrin, has been described by Lukashev et al.22 Each chimeric construct is expressed under the control of the mouse metallothionein I promoter, which can be induced by addition of ZnSO4 to the growth medium. PC3 cells were electroporated using a Genepulser apparatus set at 250 V and 900 µF using either 10 or 30 µg of the Ch1ß1A or the Chß1C chimera, respectively. Hygromycin-resistant cells were selected using growth medium containing 0.2 mg/ml hygromycin B (Boehringer Mannheim). Hygromycin-resistant colonies were pooled, and the resultant population was analyzed for cell surface expression of each chimeric construct by fluorescence-activated cell sorting using rat monoclonal antibody to mouse CD4. As negative control antibody, isotype-matched rat IgG was used. Stable transfectants were maintained in growth medium containing 0.2 mg/ml hygromycin B.
Cell Proliferation Assay
PC3 stable transfectants were starved for 24 hours in serum-free medium. Surface expression of chimeric constructs was induced by treating the cells with 75 µmol/L ZnSO4 (Sigma) in growth medium for 6 hours at 37°C. Cells were then detached with 0.05% trypsin/0.53 mmol/L ethylenediaminetetraacetic acid (Life Technologies), washed three times in serum-free medium, and analyzed by fluorescence-activated cell sorting using rat monoclonal antibody to mouse CD4 or isotype-matched rat IgG. Positive cells expressing comparable levels of the chimeric constructs were sorted using FACStar (Becton Dickinson). After sorting, the cell transfectants (15 x 103) were resuspended in serum-free medium and added to 96-well microtiter plates (ICN Biomedicals, Costa Mesa, CA) coated with 3 µg/ml fibronectin to ensure comparable attachment. After 1 hour and 30 minutes at 37°C, cells were washed twice, incubated either in the absence or in the presence of 10% FCS for 18 hours at 37°C, and pulsed with 1 µCi[3H]thymidine/well (5.0 Ci/mmol; Amersham Life Sciences, Arlington Heights, IL) during the last 3 hours of the 18-hour culture. Thymidine incorporation was evaluated as described.12 In each experiment, duplicate or triplicate observations were performed, and the values are reported as mean ± standard error (SE). In parallel, in each experiment, the number of attached cells was evaluated after fixing and staining the cells with 0.5% crystal violet (Sigma), as described.29 The results were evaluated using a 630-nm wavelength filter in a Titertek Multiskan Bichromatic enzyme-linked immunosorbent assay reader (ICN). Group differences were compared using one-way analysis of variance followed by Bonferroni post hoc contrast.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Immunohistochemical analysis of adult and fetal human tissue
specimens was performed using an affinity-purified antibody to
ß1C; ß1C was detected in a subset of
epithelial cells. Specifically, simple or pseudostratified epithelia in
prostate, liver, gallbladder, lung, and kidney were positive for
ß1C (see Figures 1 through 6 and 8
, and data not shown)
in all the analyzed cases, whereas stratified squamous epithelium of
the epidermis was negative (Figure 1E)
.
In the prostate, ß1C was preferentially expressed in the
luminal glandular epithelium (Figure 1A)
. In contrast, the basal
epithelium, composed of proliferating cells30
identified by
an antibody to basal cell-specific cytokeratins 1, 5, 10, and 14
(Figure 1D)
, did not reveal detectable levels of ß1C
(Figure 1A)
. Similarly, ß1C immunoreactivity was
undetectable in stromal and endothelial cells in all the analyzed
tissues (see Figures 1 through 6
). The specificity of the
immunoreactivity of the antibody to ß1C (Figure 2A)
was confirmed by 1) inhibition by the
ß1C peptide, 785 to 808, used as the antigen (Figure 2B)
,
2) immunoblotting of detergent lysate of prostate tissue (Figure 2E)
,
and 3) immunoprecipitation of surface-iodinated BPH-1 luminal
epithelial cells (Figure 2F)
. The immunoblotting and
immunoprecipitation both revealed bands with an electrophoretic
mobility similar to that of ß1A (Figure 2, E and F)
.12
|
|
|
and not shown), whereas it was absent
in budding glands (Figure 4A)
. Cytokeratin 8 and 18 staining was used
for the identification of prostate glands31
that had not
differentiated completely and lacked a lumen (Figure 4C)
. The basal
epithelium, identified using an antibody to basal cell-specific
cytokeratins 1, 5, 10, and 14 (Figure 3D)
, did not contain detectable
ß1C (Figure 3A)
. The specificity of ß1C
immunoreactivity (Figure 3A)
was confirmed by inhibitions with the
ß1C peptide used as the antigen (Figure 3C)
. These
staining results suggest that ß1C is a marker of prostate
epithelial cell differentiation.
|
, which stained at the apical, basal, and lateral cell surfaces;
no ß1C was found in hepatocytes (Figure 5C)
. A similar
staining pattern showing expression in bile ducts, but not in
hepatocytes, was observed in 16, 17, and 19 week-gestation fetal livers
(Figure 5, G and H
, and not shown). The specificity of the staining was
confirmed by inhibition with the ß1C peptide (Figure 5E)
.
The control of ß1C expression appears to be independent
of ß1A, as the two variants are differentially regulated
in a cell type-specific manner. In contrast to ß1C, which
has a distribution restricted to biliary epithelium (Figure 6, A and B)
, the ß1A
variant was found in bile duct epithelium (Figure 6D)
; hepatic
vasculature (Figure 6C
and not shown); and, weakly, in hepatocytes, as
shown previously.32
|
|
Because ß1C has a powerful growth-inhibitory
activity in fibroblasts12,13
and because, as shown above,
it is predominantly found in nonproliferative epithelium, we explored
the possibility that ß1C would have a causal role in
inhibiting epithelial cell proliferation. Chimeric constructs
containing either ß1C (Chß1C; Figure 7, A and B
) or ß1A
(Ch1ß1A; Figure 7, C and D
) cytoplasmic tails under the
control of an inducible promoter were stably expressed in PC3 prostate
cancer cells. Serum-starved stable transfectants were incubated with
ZnSO4 (Figure 7, A and C)
to induce surface expression of
the chimeric proteins and sorted by fluorescence-activated cell sorting
to isolate positive cells expressing comparable levels of
Chß1C or Ch1ß1A. The sorted cells were
compared in proliferation assays on fibronectin (Figure 7E)
. Under the
experimental conditions used, adhesion to fibronectin was not affected
by the expression of either Chß1C or Ch1ß1A
(not shown). However, the expression of ß1C completely
inhibited thymidine incorporation in response to serum in the cells,
whereas ß1A had no effect.
|
, which was observed in normal
and hyperplastic prostatic glandular epithelia of all the analyzed
cases, was selectively downregulated in regenerative areas displaying
hyperchromatic nuclei and cellular crowding (Figure 8, C and D)
.
ß1A was always detected in both the basal and luminal
layers (Figure 8B
, and not shown). These results suggest a unique role
for ß1C in controlling epithelial cell proliferation
under normal and pathological conditions.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Our in vivo study shows a selective expression of ß1C in a subset of epithelial cells that does overlap with the expression of ß1A. This analysis of ß1C distribution was made possible by the availability of highly specific antibodies to the ß1C cytoplasmic domain and emphasizes the importance of using antibodies specific to the different variants when analyzing ß1 distribution. The ß1C distribution is unique among the ß1 variants. The ß1B isoform has been found to be restricted to skin and liver tissues,33 whereas the ß1D subunit is expressed in striated muscles, where it replaces ß1A19,34 in a developmentally regulated manner.35
Functional differences have been described for the ß1 variants. ß1D is functionally similar to ß1A, in that both are localized at focal contacts and can stimulate focal adhesion kinase activation,19,34 whereas ß1B inhibits it.15 This suggests that modulation of splicing patterns of ß1 mRNA may provide an accessory mechanism to regulate signaling pathways initiated by integrins. A recent study shows that the ability of ß1C to inhibit cell proliferation is shared by ß1D36; however, the downstream signaling pathways by these variants are unknown and may be different.
The mechanisms that regulate epithelial cell proliferation and differentiation in developing embryonal and fetal tissue are largely unknown. In general terms during organogenesis, a proliferative phase is followed by one in which the cells progressively differentiate and acquire the phenotype required for their highly specialized functions.37 The results of our study indicate that ß1C is expressed at the transition between these phases and suggest that its expression might be an important mechanism by which cellular proliferation is inhibited in differentiating fetal epithelia in specific organs.
Altered expression of integrins in epithelial cells has been shown to generate pathological phenotypes.32 Abnormal expression of ß1 in the suprabasal epidermal layers of the skin causes epidermal hyperproliferation, as shown in studies performed using transgenic mice, thus confirming that incorrect integrin distribution can be a trigger for increased in vivo growth rate.38,39 In normal skin, ß1 integrins are found only in the basal layer of the epidermis, whereas during wound healing and psoriasis, which are associated with hyperproliferation, they also appear in the suprabasal layers of the skin.40 Integrin expression affects the growth of tumor cells in vitro and tumor growth and metastasis in vivo.41-50 On the basis of the different effects of ß1C and ß1A on cell proliferation, we would predict that replacement of the ß1C with ß1A could also lead to abnormal proliferation in epithelium.
The expression of ß1C completely inhibited thymidine incorporation of prostate cancer epithelial cells in response to serum. Consistent with these in vitro results, ß1C appeared to be downregulated in prostate glands that exhibit regenerative features in benign hyperplastic epithelium. These observations, and previous findings showing that ß1C is downregulated in prostate cancer,26 suggest that the loss of ß1C expression may contribute malignant progression in prostate cells.
|
Acknowledgements |
|---|
|
Footnotes |
|---|
Supported in part by National Institutes of Health Grants CA 71870 and DK 52670, Donaghue Medical Research Foundation grant 95-006, a Milheim Foundation Award (to LRL), a Donaghue Medical Research Foundation Fellowship Award (to MF), and a fellowship from the American-Italian Cancer Foundation (to MM).
MF and MM contributed equally to this report.
MM's present address: Department of Pathology, European Institute of Oncology, Milan, Italy.
Accepted for publication July 10, 1998.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
6 integrin subunit. J Biol Chem 1998, 273:5903-59072ß1 integrin as a laminin receptor. J Cell Biol 1989, 109:2455-24627 and ß1D integrin during mouse heart and skeletal muscle development. Cell Adhes Commun 1998, 5:193-205[Medline]
6 and 2 very late antigen integrins in the normal, hyperplastic, and neoplastic prostate: simultaneous demonstration of cell surface receptors and their extracellular ligands. Hum Pathol 1993, 24:243-248[Medline]
6-integrins in the normal and neoplastic prostate. Am J Pathol 1994, 145:167-174[Abstract]
6ß1 promotes the survival of metastatic human breast carcinoma cells in mice. Am J Pathol 1997, 151:1191-1197[Abstract]
6ß4 in the invasive properties of colorectal carcinoma cells. Cancer Res 1996, 56:4811-4819This article has been cited by other articles:
 |
|
 |
|
  H. L. Goel, L. Moro, M. King, N. Teider, M. Centrella, T. L. McCarthy, M. Holgado-Madruga, A. J. Wong, E. Marra, and L. R. Languino {beta}1 Integrins Modulate Cell Adhesion by Regulating Insulin-Like Growth Factor-II Levels in the Microenvironment Cancer Res., January 1, 2006; 66(1): 331 - 342. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. L. Goel, M. Breen, J. Zhang, I. Das, S. Aznavoorian-Cheshire, N. M. Greenberg, A. Elgavish, and L. R. Languino {beta}1A Integrin Expression Is Required for Type 1 Insulin-Like Growth Factor Receptor Mitogenic and Transforming Activities and Localization to Focal Contacts Cancer Res., August 1, 2005; 65(15): 6692 - 6700. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Moro, E. Perlino, E. Marra, L. R. Languino, and M. Greco Regulation of {beta}1C and {beta}1A Integrin Expression in Prostate Carcinoma Cells J. Biol. Chem., January 16, 2004; 279(3): 1692 - 1702. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Fornaro, J. Plescia, S. Chheang, G. Tallini, Y.-M. Zhu, M. King, D. C. Altieri, and L. R. Languino Fibronectin Protects Prostate Cancer Cells from Tumor Necrosis Factor-{alpha}-induced Apoptosis via the AKT/Survivin Pathway J. Biol. Chem., December 12, 2003; 278(50): 50402 - 50411. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Lovecchio, E. Maiorano, R. A. Vacca, G. Loverro, M. Fanelli, L. Resta, S. Stefanelli, L. Selvaggi, E. Marra, and E. Perlino {beta}1C Integrin Expression in Human Endometrial Proliferative Diseases Am. J. Pathol., December 1, 2003; 163(6): 2543 - 2553. [Abstract] [Full Text]
|
|
|
|
 |
|
 R. A. Vacca, E. Marra, G. Loverro, E. Maiorano, A. Napoli, M. Lovecchio, L. Selvaggi, and E. Perlino Differential Expression of {beta}1c Integrin Messenger Ribonucleic Acid and Protein Levels in Human Endometrium and Decidua during the Menstrual Cycle and Pregnancy J. Clin. Endocrinol. Metab., February 1, 2003; 88(2): 720 - 729. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Miralem and H. K. Avraham Extracellular Matrix Enhances Heregulin-Dependent BRCA1 Phosphorylation and Suppresses BRCA1 Expression through Its C Terminus Mol. Cell. Biol., January 15, 2003; 23(2): 579 - 593. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. Perlino, M. Lovecchio, R. A. Vacca, M. Fornaro, L. Moro, P. Ditonno, M. Battaglia, F. P. Selvaggi, M. G. Mastropasqua, P. Bufo, et al. Regulation of mRNA and Protein Levels of {beta}1 Integrin Variants in Human Prostate Carcinoma Am. J. Pathol., November 1, 2000; 157(5): 1727 - 1734. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Fornaro, C. A. Steger, A. M. Bennett, J. J. Wu, and L. R. Languino Differential Role of beta 1C and beta 1A Integrin Cytoplasmic Variants in Modulating Focal Adhesion Kinase, Protein Kinase B/AKT, and Ras/Mitogen-activated Protein Kinase Pathways Mol. Biol. Cell, July 1, 2000; 11(7): 2235 - 2249. [Abstract] [Full Text]
|
|
|
|
|
|
M. Manzotti, P. Dell’Orto, P. Maisonneuve, M. Fornaro, L. R. Languino, and G. Viale Down-Regulation of {beta}1C Integrin in Breast Carcinomas Correlates with High Proliferative Fraction, High Histological Grade, and Larger Size Am. J. Pathol., January 1, 2000; 156(1): 169 - 174. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G Svineng and S Johansson Integrin subunits (beta)1C-1 and (beta)1C-2 expressed in GD25T cells are retained and degraded intracellularly rather than localised to the cell surface J. Cell Sci., January 12, 1999; 112(24): 4751 - 4761. [Abstract] [PDF]
|
|
|
|
|
|
A. L. Jannuzi, T. A. Bunch, M. C. Brabant, S. W. Miller, L. Mukai, M. Zavortink, and D. L. Brower Disruption of C-Terminal Cytoplasmic Domain of beta PS Integrin Subunit Has Dominant Negative Properties in Developing Drosophila Mol. Biol. Cell, April 1, 2002; 13(4): 1352 - 1365. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
