| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
* Department of Bacteriology and Immunology, Haartman Institute, University of Helsinki, Helsinki, Finland;
Division of Immunology, HUSLAB Helsinki University Central Hospital Laboratory, Helsinki, Finland;
Department of Ophthalmology, University of Helsinki, Helsinki, Finland;
Center for the Study of Macular Degeneration, Neuroscience Research Institute, University of California, Santa Barbara, CA 93106;
¶ Department of Medical Genetics, University of Helsinki, Helsinki, Finland;
|| Laboratory of Molecular Genetics, Helsinki University Hospital, Helsinki, Finland; and
# Department of Ophthalmology and Visual Sciences, Center for Macular Degeneration, University of Iowa, Iowa City, IA 52240
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
C, the tyrosine in position 402 in SCR7 of FH is substituted by a histidine. Individuals homozygous for 402H (genotype CC) have a 5- to 7-fold increased risk for developing AMD than individuals homozygous for 402Y (genotype TT). Heterozygotes (genotype TC) also have an increased risk, but a less significant one (7). Complement FH promotes dissociation of the alternative pathway C3 convertase and inactivation of C3b to iC3b, thereby inhibiting complement activation. iC3b is a crucial ligand for the macrophage and dendritic cell receptors CD11a-c/18. Factor H binds to complement C3b via interaction sites in SCR domains 1–4, 12–14, and 19–20 (10). These interactions down-modulate complement activation in human plasma and on viable host cell surfaces. Receptors for FH on cell surfaces include negatively charged glycosaminoglycans, phospholipids, and sialic acids. FH has up to three distinct heparin (a model polyanion) binding sites, one of which is located within SCR7 (11, 12).
In addition to heparin and C3b, FH also binds to C-reactive protein (CRP) at two sites, one located at SCR7 and the other in a region spanning between SCR8 and SCR11 (12, 13). Once bound to phosphocholine, CRP activates the classical complement pathway by also binding C1q (14, 15). Subsequent binding of FH to CRP suppresses activation of the alternative complement pathway (13, 16).
An alternatively spliced product of the Cfh gene, FH-like protein 1 (FHL-1), also regulates the alternative complement pathway. It is composed of the first seven SCR domains of FH and four unique amino acids (17) and thus contains the same AMD-associated polymorphism. A group of proteins called FH-related proteins (FHRs) also exists. It consists of five proteins that are structurally similar to FH. Each FHR is a product of its own gene distinct from Cfh. FHRs consist of four to nine SCR domains that have a varying degree of homology to those of FH. The homolog of SCR7 is present in all of them except in FHR-4 (17).
Based on existing information, it is logical to speculate that a polymorphism in FH could affect its binding to surface polyanions and/or CRP, or influence the production of FH and/or FHL-1. Therefore, we sought in this investigation to explore the effects of the FH Y402H polymorphism on the FH protein function.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
We collected DNA and serum samples from 118 carefully phenotyped familial AMD cases seen in the Department of Ophthalmology (Helsinki University Hospital, Helsinki, Finland). AMD was verified and graded (large confluent drusen, central geographic atrophy, or neovascular AMD) from fundus photographs and/or angiograms from all patients. A control group that comprised of 71 patients was collected from subjects attending the same hospital for cataract operations. These subjects had no large drusen and no, or minimal, focal pigmentary abnormalities in the macula as verified by fundus photographs and clinical examination by a specialist. All patients were genotyped for the FH 402 polymorphism as described below. For CRP-FH binding studies, a random cohort of 46 AMD patients and 33 control patients was selected out of the larger cohort. The mean age of the AMD cases was 77.0 (range 58.1–92.4) and that of the control subjects was 76.6 (range 69.8–87.5). The study was approved by the Ethics Committee of the Helsinki University Eye and Ear Hospital.
Genotyping
DNA was obtained from peripheral blood samples, amplified by PCR, and sequenced using the forward primer 5'-ctttgttagtaactttagttcg-3' and the reverse primer 5'-ttagaaagacatgaacatgctagg-3' of the Cfh gene. PCR amplifications were conducted in a 50-µl volume containing 80 ng of genomic DNA, 30 pM of each primer, polymerase buffer, 10 nM of each nucleotide (dNTP), and 0.8 U of Dynazyme polymerase-enzyme (Finnzymes). Sequencing was performed using cycle sequencing with the Big Dye Terminator kit (version 3.1; Applied Biosystems) and reactions were run on an ABI 3730 capillary sequencer according to the manufacturer’s instructions.
Immunohistochemistry
Sections of human eyes were obtained, stained, and analyzed by laser scanning confocal immunofluorescence microscopy as described by Johnson et al. (18).
SDS-PAGE and immunoblotting
To determine whether the FH Y402H polymorphism influenced the molecular nature of serum FH, serum samples from 54 AMD patients and controls either homozygous for FH402Y (genotype TT) or FH402H (genotype CC) or heterozygous for the Y402H polymorphism (genotype TC) were analyzed by immunoblotting using polyclonal and monoclonal (86X and 90X) Abs directed against FH (19). Serum samples from patients and controls were separated on 10% SDS-PAGE gels under nonreducing conditions. The proteins were transferred onto a nitrocellulose membrane and nonspecific binding sites were blocked. Polyclonal goat anti-human FH (Calbiochem) or mAbs and peroxidase-conjugated donkey anti-goat IgG (Jackson ImmunoResearch Laboratories) or peroxidase-conjugated rabbit anti-mouse IgG (Jackson ImmunoResearch Laboratories) were used for the detection of FH and FHRs.
Quantification of CRP
CRP levels were determined using a highly sensitive immunoturbidimetric assay (Orion Diagnostica). CRP concentration was not available for three TT controls and one TC control.
Purification of FH402Y and FH402H
The IgG fraction of goat antiserum to human FH (Quidel) was precipitated with 18% Na2SO4. The precipitate was washed with 18% Na2SO4 in PBS (pH 7.4) and resuspended in sodium carbonate buffer. The precipitate was then coupled to CNBr-activated Sepharose 4B (Amersham Biosciences) according to the manufacturer’s instructions. Sera from 10 controls with the TT genotype were pooled for isolation of FH402Y and sera from 10 patients with the CC genotype were pooled for isolation of and FH402H. Both serum pools were incubated with Sepharose for 1 h and the column was washed with 0.5 M NaCl in PBS. The bound FH proteins were eluted with the chaotropic agent sodium thiocyanate, 3 M, and finally dialyzed against veronal-buffered saline. The purity of the proteins was checked by SDS-PAGE and both silver staining and Western blotting. The concentrations of purified proteins were determined by using absorbance measurements at 280 nm, the BCA Protein Assay (Pierce), and silver staining.
Expression of rFH SCR5–7 and construction of FH SCR5–7402H variants
rFH SCR5–7 constructs were generated as described by Jokiranta et al. (20) using specific primers for SCRs 5 and 7 and pPICZ
expression vectors. The Y402H mutation was introduced to the FH SCR5–7 sequence by using the QuikChange MultiSite Mutagenesis kit (Stratagene).
Binding of FH to CRP
Maxisorp microtiter plates (Nunc) were coated with CRP (1 µg/ml; Calbiochem), whose purity was confirmed with silver staining. After blocking, sera (1/1,000) from patients and controls, purified FH proteins, or rSCR5–7 variants were added. The plates were incubated for 1 h at 37°C and washed. Polyclonal goat anti-human FH (1/5,000) was added. Peroxidase-conjugated donkey anti-goat IgG diluted 1/10,000 was used as the secondary Ab. Chromogenic substrate and absorbance measurement at 492 nm were used for detection and results indicated as direct OD values. All the FH-CRP binding experiments were repeated at least twice on duplicate samples.
Binding of FH to heparin
Binding of purified FH proteins (FH402Y and FH402H) and respective SCR5–7 fragments to heparin was analyzed using heparin affinity chromatography in an HPLC system (LaChrom L-7100; Hitachi). Ten micrograms of proteins was diluted in 1/2x PBS and applied to a heparin-Sepharose affinity column (HiTrap; Amersham Biosciences) at a flow rate of 0.5 ml/min. The column was extensively washed with 0.5x PBS and the bound proteins were eluted using a linear salt gradient ranging from 75 to 500 mM NaCl, in a total volume of 10 ml and at a flow rate of 0.5 ml/min.
Molecular modeling
The schematic molecular display model depicting binding between FH and CRP presented in Fig. 4 is based upon the crystal structure of CRP (21) and 10 copies of nuclear magnetic resonance structures of the FH SCR15–16 domain pair as display units (22). The model was created using InsightII 2000 (Accelrys).
|
Statistical analyses between more than two groups were conducted using ANOVA. Post hoc comparisons between individual groups were done using the Games-Howell test as the variances of different groups were not homogeneous. Statistical analyses between two groups were done using the two-tailed Student t test.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
FH proteins in sera of individuals with the three different 402 genotypes (TT, TC, or CC) showed similar migration patterns in SDS-polyacrylamide gels (Fig. 1A). In some serum samples, an extra, more slowly migrating protein band was observed but its presence did not correlate with the FH genotype or disease status. The identity of the band is unknown, but it is commonly seen in FH serum blots. Binding of the two full-length FH variants, FH402Y derived from controls and FH402H derived from patients, to heparin was assessed using heparin-Sepharose affinity chromatography. Both protein variants eluted from the column at the same ionic strength (Fig. 1B), indicating that the Y402H variation does not markedly affect binding of FH to heparin. Similar results were obtained using recombinant fragments of FH SCR5–7 containing either the 402Y or 402H residues.
|
Drusen have been demonstrated to contain complement components. With regard to possible interactions between FH and CRP their expression was analyzed by confocal immunofluorescence microscopy analysis. As shown in two representative cases (Fig. 2), FH was present throughout the drusen in individuals with either the TT or CC genotype. Intriguingly, however, CRP was present in small spheroid particles inside the drusen (Fig. 2, upper panel).
|
To analyze binding of the different variants of FH to CRP, we set up an ELISA by coating plate wells with CRP and testing binding of serum-derived FH, purified FH, and rFH SCR5–7 fragments to CRP. Significant differences in binding to CRP were found to exist between the different groups (TT controls, TC controls, CC controls, TT patients, TC patients, and CC patients) (F = 15.3, p = 2.9 x 10–10). FH in the sera of AMD patients homozygous for the 402H (CC genotype) exhibited strongly reduced binding to CRP (OD492: 0.369 ± 0.102, n = 20), as compared with patients with the TT genotype (OD492: 0.532 ± 0.102, n = 16, p = 4.9 x 10–4) (Fig. 3A). A similar difference was observed using sera from healthy controls indicating that the functional difference in CRP binding is genotype associated (CC controls: OD492: 0.324 ± 0.098, n = 9, TT controls: OD492: 0.697 ± 0.199, n = 14, p = 1.0 x 10–4) (Fig. 3A). Individuals heterozygous for the C allele (genotype TC) also exhibited reduced binding of serum FH to CRP (Fig. 3A). When data from the patient and control groups were pooled according to genotype, a significant difference in FH binding to CRP was observed (F = 27.4, p = 1.1 x 10–9). This difference was most marked between the CC (OD492: 0.355 ± 0.101, n = 29) and TT genotypes (OD492: 0.609 ± 0.173, n = 30, p = 3.9 x 10–8) and between the TC (OD492: 0.499 ± 0.100, n = 20) and TT genotypes (p = 4.2 x 10–4) but the difference between CC and TC genotypes was also significant (p = 6.8 x 10–3). When AMD patient data were pooled regardless of genotype (OD492: 0.445 ± 0.129, n = 46) and compared with pooled control data (OD492: 0.518 ± 0.214, n = 33), no significant difference in FH binding to CRP was observed (p = 0.062). Similar analyses using full-length FH variants (FH402Y and FH402H) purified from sera were performed to exclude any interfering factors in serum samples. Binding of purified FH402H protein to CRP (OD492: 0.470 ± 0.033) was significantly weaker than that of the FH402Y protein (OD492: 0.668 ± 0.103, p = 0.034) (Fig. 3B). An even more marked difference was observed between the two recombinant fragments of FH in binding to CRP than using the full-length proteins (Fig. 3C). By measuring FH concentrations of the samples, we excluded varying FH levels as the cause for these results (data not shown).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
We observed that the AMD-associated variant of FH, FH402H, has a lower binding affinity for CRP than the FH402Y variant. Binding of FH to CRP is a critical interaction through which CRP participates in the regulation of complement activation. Drusen in the eyes examined contained spheroid-like particles, which in a TT-genotyped, but not in a CC-genotyped, individual showed costaining for CRP and FH (Fig. 2). In a recent study, CRP was found to be more abundantly expressed in the eyes of AMD patients with FH402H than those with FH402Y (18). CRP, a plate-like pentamer, binds to surface phosphocholine or cholesterol in a Ca2+-dependent manner with each of its five subunits (23). C1q binds to a different region of the pentameric plate via its globular domains (24). Through these interactions, CRP recognizes a target, activates the classical complement pathway, and restricts activation of the complement cascade to the C3 level by binding FH. The domains of FH that mediate binding to CRP lie within SCR7 and within a region that spans SCRs 8–11 (13). A model of the CRP-FH interaction is depicted in Fig. 4.
Our experiments using recombinant fragments comprised of FH SCR5–7 show that 402Y in SCR7 is an important amino acid for the binding of FH to CRP. It could be directly involved in making contact with the CRP molecule on the surface of the SCR7 domain. The Y402H change might thus disrupt the interaction surface with CRP or destabilize the interdomain interface between SCR7 and SCR8 and lead to a conformational change in the whole FH molecule (25, 26). We postulate that an impairment of the FH-CRP interaction leads to an inadequate clearance of eye pigment derived lipofuscin or damaged tissue, debris accumulation, and inflammation (13). This would be the consequence of an inability to locally control activation of the alternative complement pathway (27).
Due to a second CRP-binding site in SCR8–11, the full-length FH seems to be able to somewhat compensate for the reduced binding of SCR7 containing 402H to CRP. This could explain the smaller differences in CRP binding between full-length FH variants than between the rSCR5–7 proteins that lack the second binding site (Fig. 3, B vs C). Because the SCR8–11 binding site is not present in FHL-1, the direct effect of 402Y in binding to CRP might be significant in FHL-1 and important for its function as a local regulator of the alternative complement pathway. FHL-1 consist of SCRs 1–7. Thus, the amino acid 402 in the most C-terminal SCR domain of FHL-1 is more likely directly involved in binding to CRP. Although FHL-1 levels are only about one-tenth of those of FH in plasma, its concentrations can be considerably higher locally and it can be important in controlling local inflammatory processes (17). Based on our recombinant protein data, this function is influenced by the Y402H polymorphism. Another interesting finding is that differences between the CRP-binding affinity of the two FH variants seem to be bigger in assays with serum samples than in assays with purified FH. This could, at least in part, be due to serum FHRs that possibly compete with FH for binding to CRP or C3b. In the case of FHR-1 and FHR-3 deficiency, that has been proposed to be protective against AMD (28, 29), the FH-CRP interaction might be stronger because of a lack of competition by these FHRs for binding to CRP.
The targets for CRP in macular lesions are unknown. Cholesterol has been demonstrated in drusen (30) and is a known ligand for CRP (23). This interaction is highly specific, because not even the closely related epicholesterol is capable of binding to CRP. Other candidate receptors for CRP within the macula might include products of retinal pigment catabolism and/or RPE-derived cell components or protein aggregates in the form of spheroid-like structures as seen in the drusen (Fig. 2).
The production of CRP increases during episodes of inflammation and/or tissue injury (31). When bound to one of its ligands on a surface CRP binds C1q and activates the classical complement pathway. CRP also binds FH, which serves as a cofactor for factor I to inactivate accumulating C3b molecules. The formed iC3b molecules subsequently function as high-affinity ligands for the dendritic cell and macrophage complement receptor CD11b/18, which mediates noninflammatory phagocytosis of opsonized tissue components (13, 32). The same process down-modulates alternative pathway amplification and reduces inflammation and formation of MAC by preventing activation of C5 to the C5a anaphylatoxin and C5b. Formation of iC3b would be less efficient in individuals with the FH402H variant. As such, the inflammation suppressing effects of FH and CRP would be less effective. In consequence, this would lead to a less controlled inflammatory reaction and more tissue damage.
The episodic nature of CRP elevation and related dysregulated complement activation in genetically predisposed individuals is compatible with our knowledge of AMD histopathology. We propose that excessive tissue injury develops, over time, as a result of episodic inflammation and deposition of complement. CRP would normally regulate and target the inflammatory process. This function would be disturbed in individuals with FH402H. CRP levels could become elevated because of infections and/or tissue damage at distant sites and consequences in the eye could be an indirect side effect of a process elsewhere.
In our Finnish study cohort, the presence of the CC genotype carried a 9.7-fold increased risk for AMD (33). Although the binding of FH to CRP is diminished in patients with the CC genotype, not all AMD patients possess this genotype. Other factors are thus also likely to be involved in the pathogenesis of AMD. These might be other polymorphisms in FH itself (e.g., I62V) or in other complement components. This is exemplified by recent findings that variations in genes coding for complement components factor B and C2 are associated with AMD (34), and that a combined FHR-1 and FHR-3 deletion is protective against the disease (28, 29).
In some studies, elevated levels of CRP have been associated with the progression of AMD (35, 36). However, in our analysis the difference in CRP levels between patients and controls (p = 0.22) or different genotypes (F = 0.51, p = 0.61) was not significant. This result excludes the possibility that varying levels of CRP are the cause for the differences in FH binding to CRP between the different genotypes in this study.
Binding of FH to polyanions is important in the discrimination between alternative pathway activators and nonactivators (37). FH binds to glycosaminoglycans, including heparan sulfate, via its SCR7, SCR9, SCR13, and SCR19–20 domains. The data from this study suggest that compared with CRP binding, it is less likely that a difference in glycosaminoglycan binding between the two proteins is involved in AMD pathogenesis.
In conclusion, we have shown that the polymorphic FH402H variant has a reduced capacity to bind CRP. This could result in excessive inflammation induced by distinct drusen-associated structures (shown in Fig. 2) in individuals who carry this FH variant and develop AMD. Systemic implications of the reduced binding of FH to CRP are also possible and remain to be investigated.
|
Acknowledgments |
|---|
|
Disclosures |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
|
Footnotes |
|---|
1 This work was supported by the Academy of Finland, Sigrid Jusélius Foundation, Helsinki University Central Hospital Funds (EVO), Päivikki and Sakari Sohlberg Foundation, Finska Läkaresällskapet, The Eye and Tissue Bank Foundation, The Eye Foundation (Helsinki, Finland), Kidneeds Foundation, the National Institutes of Health (EY11515), the American Macular Degeneration Foundation, and the Eye Research Institute. 
2 Address correspondence and reprint requests to Dr. Seppo Meri, Department of Bacteriology and Immunology, University of Helsinki, P.O. Box 21, Haartman Institute (Haartmaninkatu 3), FI-00014, Helsinki, Finland. E-mail address: seppo.meri{at}helsinki.fi
3 Abbreviations used in this paper: AMD, age-related macular degeneration; RPE, retinal pigmented epithelium; MAC, membrane attack complex; FH, factor H; SCR, short consensus repeat; CRP, C-reactive protein; FHL-1, complement FH-like protein 1; FHR, FH-related protein.
Received for publication September 7, 2006. Accepted for publication December 20, 2006.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
-OH group of cholesterol in LDL particles. Biochem. Biophys. Res. Commun. 329: 1208-1216. [Medline]This article has been cited by other articles:
|
|
 |
|
  J. Zhou, S. R. Kim, B. S. Westlund, and J. R. Sparrow Complement Activation by Bisretinoid Constituents of RPE Lipofuscin Invest. Ophthalmol. Vis. Sci., March 1, 2009; 50(3): 1392 - 1399. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Deban, H. Jarva, M. J. Lehtinen, B. Bottazzi, A. Bastone, A. Doni, T. S. Jokiranta, A. Mantovani, and S. Meri Binding of the Long Pentraxin PTX3 to Factor H: Interacting Domains and Function in the Regulation of Complement Activation J. Immunol., December 15, 2008; 181(12): 8433 - 8440. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Friberg, P. Carlson, E. Kentala, P. S. Mattila, P. Kuusela, S. Meri, and H. Jarva Factor H Binding as a Complement Evasion Mechanism for an Anaerobic Pathogen, Fusobacterium necrophorum J. Immunol., December 15, 2008; 181(12): 8624 - 8632. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Hakobyan, C. L. Harris, C. W. van den Berg, M. C. Fernandez-Alonso, E. G. de Jorge, S. R. de Cordoba, G. Rivas, P. Mangione, M. B. Pepys, and B. P. Morgan Complement Factor H Binds to Denatured Rather than to Native Pentameric C-reactive Protein J. Biol. Chem., November 7, 2008; 283(45): 30451 - 30460. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. C. Jacobson and J. H. Weis Comparative Functional Evolution of Human and Mouse CR1 and CR2 J. Immunol., September 1, 2008; 181(5): 2953 - 2959. [Full Text] [PDF]
|
|
|
|
|
|
S. Hakobyan, C. L. Harris, A. Tortajada, E. Goicochea de Jorge, A. Garcia-Layana, P. Fernandez-Robredo, S. Rodriguez de Cordoba, and B. P. Morgan Measurement of Factor H Variants in Plasma Using Variant-Specific Monoclonal Antibodies: Application to Assessing Risk of Age-Related Macular Degeneration Invest. Ophthalmol. Vis. Sci., May 1, 2008; 49(5): 1983 - 1990. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. J. Ormsby, S. Ranganathan, J. C. Tong, K. M. Griggs, D. P. Dimasi, A. W. Hewitt, K. P. Burdon, J. E. Craig, J. Hoh, and D. L. Gordon Functional and Structural Implications of the Complement Factor H Y402H Polymorphism Associated with Age-Related Macular Degeneration Invest. Ophthalmol. Vis. Sci., May 1, 2008; 49(5): 1763 - 1770. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. J.H. Smith, J. Alexander, P. N. Barlow, M. Botto, T. L. Cassavant, H. T. Cook, S. R. de Cordoba, G. S. Hageman, T. S. Jokiranta, W. J. Kimberling, et al. New Approaches to the Treatment of Dense Deposit Disease J. Am. Soc. Nephrol., September 1, 2007; 18(9): 2447 - 2456. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Z. Wu, T. W. Lauer, A. Sick, S. F. Hackett, and P. A. Campochiaro Oxidative Stress Modulates Complement Factor H Expression in Retinal Pigmented Epithelial Cells by Acetylation of FOXO3 J. Biol. Chem., August 3, 2007; 282(31): 22414 - 22425. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. P. Herbert, J. A. Deakin, C. Q. Schmidt, B. S. Blaum, C. Egan, V. P. Ferreira, M. K. Pangburn, M. Lyon, D. Uhrin, and P. N. Barlow Structure Shows That a Glycosaminoglycan and Protein Recognition Site in Factor H Is Perturbed by Age-related Macular Degeneration-linked Single Nucleotide Polymorphism J. Biol. Chem., June 29, 2007; 282(26): 18960 - 18968. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
