| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Articles |
a Department of Anatomy and Reproductive Biology, University of Hawaii School of Medicine, Honolulu, Hawaii 96822
b Signalling Program, Babraham Institute, Cambridge, CB2 4AT, United Kingdom
c Department of Biochemistry, University of Bristol, School of Medical Sciences, Bristol, BS8 1TD, United Kingdom
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Sperm-associated proteins can be categorized according to whether they are 1) absorbed from epididymal secretions [8], 2) released from spermatozoa during epididymal transit [9], or 3) processed by endoproteolysis [10, 11]. Roles in egg-binding/penetration have been proposed for several sperm antigens that are acquired either de novo in the epididymis, or else undergo processing during maturation [2, 6, 10, 12].
Epididymal spermatozoa are largely transcriptionally and translationally quiescent. This suggests that most proteins important in maturation are secreted by the epithelium of the epididymis (as inferred from examples in category 1, above). Since secretory proteins are often encoded by abundant transcripts, information about the mechanisms regulating sperm maturation may be gained by isolating and characterizing cDNA clones corresponding to abundant species of epididymal mRNA. Accordingly, we employed a strategy to isolate clones matching abundant epididymal mRNAs, and used it to identify a novel, major protein, designated cynomolgus monkey-epididymal secretory protein 13.2, cy-ESP13.2, from the epididymis of the cynomolgus monkey, Macaca fascicularis. Anti-peptide antibodies were raised against predicted cy-ESP13.2 to investigate cy-ESP13.2 protein expression. We discuss clues to the role of ESP13.2 from the precedents of related proteins.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Restriction endonucleases, polynucleotide kinase, ribonuclease (RNase)-free deoxyribonuclease (DNase), AMV reverse transcriptase, oligo-(dT)12–18, deoxynucleotides, and pre-cut cloning vector DNAs were obtained from Pharmacia Ltd. (Milton Keynes, Bucks, UK). Hybond N nylon membrane was purchased from Amersham International (Bucks, UK). [
-32P]dCTP (> 3000 Ci/mmol) was from NEN Life Science Products (Brussels, Belgium). Nitrocellulose filters and micro-dialysis membranes were from Schleicher and Schull (Keene, NH) and Millipore (Bedford, MA), respectively. Calf intestinal alkaline phosphatase (molecular biology grade), Taq DNA polymerase, and proteinase K were from Boehringer Mannheim GmbH (Lewes, UK), and low-melting-temperature agarose from Gibco-BRL (Paisley, Scotland). All other chemicals were of the purest grade available. Oligonucleotides for sequencing and hybridization were synthesized on a DuPont (Wilmington, DE) Coder 300 synthesizer using phosphoramidite chemistry and were used without subsequent purification. Fresh tissue samples were obtained from adult male rats (Wistar strain), mice (CFLP), adult (5–6 yr old) cynomolgus monkeys (Macaca fascicularis), and (with consent) patients (35 and 46 yr old) undergoing elective resection of testicular carcinoma. Human semen was obtained from the Bourn Hall infertility clinic, Cambridge, UK. Acutely isolated samples were snap-frozen in liquid N2.
RNA Isolation and Northern Blot Analysis
Total RNA was isolated from approximately 0.5 g frozen tissues using an SDS-proteinase K-based method essentially as described previously [13]. Where appropriate, poly(A)-containing RNA was enriched from 1.5–3 mg of total RNA. Total RNA or poly(A)-enriched RNA was fractionated by electrophoresis, blotted onto a Hybond-N nylon membrane, hybridized as described previously [13]. All Northern blots were reprobed with the mouse actin cDNA insert of pAM91 [14] to confirm equivalent track loadings and the integrity of the RNA preparations.
Cloning of cDNA Corresponding to Abundant Macaque Epididymal Transcripts: Isolation and Sequence Analysis of cy-ESP13.2 cDNA
Approximately 8 x 103 macaque epididymal cDNA clones [15] were probed with high-specific-activity 32P synthesized from 1 µg M. fascicularis poly(A)-enriched epididymal RNA. Hybridization was for 24 h under conditions of high stringency (6-strength SSC, 65°C; single-strength SSC: 0.15 M sodium chloride, 0.015 M trisodium citrate, pH 7.0) and included 1 µg/ml competitor poly(A)n. Filters were then washed at high stringency (0.1-strength SSC, 0.5% [w:v] SDS, 65°C) to remove unbound label; autoradiography followed at -70°C for 5 days. Strongly positive clones were challenged at high stringency with a probe derived from total human genomic DNA. Hybridizing clones must have contained highly repetitive elements that probably originated from contaminating genomic DNA during cDNA library construction, and they were therefore eliminated. Remaining cDNA clones were grouped according to their pattern of cross-hybridization and (partial) sequence determination. One family of cross-hybridizing clones (designated pcy-ESP13.2) was selected for further analysis. An oligonucleotide probe corresponding to a region near the 5' end of the cy-ESP13.2 contig was used to rescreen the M. fascicularis epididymal cDNA library. Twelve nonsibling clones were isolated in this way, and the sequences of the two longest (pcy-ESP13.2-a, b) were determined on both strands.
Macaque sequences were used to search the EMBL (version 56.0) nucleotide sequence database using the program Tfasta as implemented in the GCG package (version 9.0; Genetics Computer Group, Madison, WI). The M. fascicularis ESP13.2 (cy-ESP13.2) nucleotide sequence reported here has been deposited in the EMBL database (Accession Number AJ236909).
Analysis of Human ESP13.2 by Reverse Transcription (RT)-Polymerase Chain Reaction (PCR)
Approximately 1 µg total human epididymal RNA was used to program cDNA synthesis in a 100-µl reaction containing oligo(dT)12–18 (50 µg/ml); dCTP, dGTP, dATP, and dTTP (each at 500 µM); 20 mM dithiothreitol (DTT); Avian Myeloblastosis Virus (AMV) reverse transcriptase reaction buffer (AMV reverse transcriptase reaction buffer is 50 mM Tris-Cl, 50 mM KCl, and 6 mM MgCl2; pH 8.2 at 42°C); and 20 U AMV reverse transcriptase. The reaction was at 42°C for 1 h, followed by deproteination with phenol:chloroform (1:1 [v:v]) and ethanol precipitation in the presence of 2 M ammonium acetate. Pelleted cDNA was washed briefly in 80% ethanol, dried, and resuspended in 20 µl H2O. Of this, 2.5 µl was used as template in a PCR reaction containing either the primers #1 (5'AGAACCCACTGCCTCCTGATG3') and #2 (5'GAGACAGAGGCTGGAATGTTCA3'), corresponding to nucleotides (nt) 1–21 and 388–409, respectively, of cy-ESP13.2 cDNA (Fig. 1A), or #3 (5'TTCAACAGTAACAGCAACAAC3'), corresponding to nt 258–278 of human ESP13.2 cDNA (Fig. 1A) and #2, with primers at 0.5 µM each. PCR parameters were 94°C, 90 sec; cool to 62°C over 90 sec; 62°C, 120 sec; 72°C, 120 sec; 30 cycles. PCR products (from at least two independent reactions in each case) were cloned into the unique SmaI site of pUC18, and their sequences were determined on both strands. The Homo sapiens ESP13.2 nucleotide sequence EMBL Accession Number is AJ236910.
|
Western Blot Analysis
Western blotting for detection of ESP13.2 was essentially as described previously [16]. Frozen tissues were homogenized in 5 vol of ice-cold 0.25 M sucrose, 2 mM EDTA, 1 mM 4-(2-aminoethyl)benzenesulfonylfluoride, HCl (AEBSF, hydrochloride) 10 mM Tris-Cl (pH 7.2) in a Teflon (DuPont)-glass Potter homogenizer. Samples were centrifuged at 1500 x g for 10 min at 5°C. Resulting postnuclear supernatants were made 1% (w:v) with respect to SDS and further clarified by centrifugation at 10 000 x g for 10 min at 5°C. Soluble proteins were fractionated by nonreducing SDS-PAGE and blotted onto polyvinylidene fluoride (PVDF) membranes by standard procedures. Immunoreactive ESP13.2 was detected on blots by incubation with a 1:100 (v:v) diluted rabbit polyclonal antiserum, followed by 1:400 (v:v) diluted peroxidase-conjugated goat, anti-rabbit IgG, and visualized with 4-chloronaphthol dye reagent. The rabbit antiserum was raised to an internal peptide (KKTCKPEEVRSEHGWVMC) predicted from the cy-ESP13.2 cDNA sequence to have a high surface probability.
Immunofluorescence Localization of ESP13.2 in Tissues and on Spermatozoa
Frozen tissues were sectioned on a cryostat at -20°C, air-dried onto pre-cleaned glass slides, and probed with anti-ESP13.2 rabbit polyclonal antiserum diluted 1:100 (v:v) in PBS-1% (w:v) BSA, followed by 1:400 (v:v) diluted fluorescein isothiocyanate (FITC)-conjugated goat, anti-rabbit IgG as described previously [16]. Sections were visualized by phase contrast epifluorescence microscopy using a Zeiss Axiophot photomicroscope (Carl Zeiss, Inc., Thornwood, NJ). Smears of spermatozoa from different levels of the macaque epididymis and from normal human semen were prepared on glass slides, air-dried, and probed with antibodies as described above.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Cross-hybridization analysis of cDNA clones representing abundant cynomolgus monkey (M. fascicularis) epididymal mRNAs revealed one family that represented 2.3% of the total number. Sequence analysis of 3 of the cross-hybridizing cDNA clones suggested that none was full-length with respect to its corresponding mRNA transcript. We therefore rescreened the epididymal cDNA library with an oligonucleotide matching the presumed 5' end of the sequence, and, of the 12 clones isolated, two were subjected to sequence analysis. The cDNA sequences of these clones were superimposable and full-length with respect to an open reading frame of 369 nt, which began with a putative initiating ATG preceded by in- and out-of-frame stop codons; the length of the compiled cDNA was 542 nt, with a potential polyadenylation/cleavage signal (5'ATTAAA3') near its 3' end (Fig. 1A). The open reading frame encoded a predicted nascent protein of 123 amino acid residues, Mr 13 200. This polypeptide sequence harbored a potentially cleavable hydrophobic signal peptide near its N terminus, suggesting that the protein passes along the secretory pathway (Fig. 1B; [17]). To reflect these features, we have designated the protein cynomolgus monkey-epididymal secretory protein 13.2, cy-ESP13.2. The sequence of deduced cy-ESP13.2 contains no consensus sites for N-linked glycosylation and predicts a protein with a pI of 10.14. In addition, 40% of its residues are either Ala, Gly, Ser, or Thr, with a 16-residue Ala-Thr-rich stretch near its C terminus.
High-stringency Northern blot analysis of total M. fascicularis epididymal RNA using a 520-nt cy-ESP13.2 cDNA probe revealed a single, abundant epididymal transcript of 690 nt that was not detected in rat epididymal RNA (data not shown); macaque liver, kidney, or testis RNAs (Fig. 2); or human epididymal RNA (Fig. 3A). Hormonal integrity of human epididymal tissue was substantiated using macaque epididymal apical protein I (EAPI) cDNA to probe corresponding human total RNA samples; EAPI expression has previously been shown to be highly androgen-sensitive (Fig. 3B) [16]. This also confirms the existence of a human EAPI ortholog. The major human EAPI transcript is similar in length to macaque EAPI mRNA (Fig. 3B). Independently sourced human samples lacked detectable ESP13.2 mRNA, suggesting either low levels of hu-ESP13.2 expression, or significant sequence disparity between cy- and hu-ESP13.2. To investigate this, we analyzed human epididymal RNA with the more sensitive technique, RT-PCR.
|
|
Cloning and Northern Blot Analysis of Human ESP13.2 mRNA
Primers #1 and #2, derived from sequences near the ends of the predicted coding region of cy-ESP13.2, were used to amplify a human epididymal ortholog of cy-ESP13.2 by RT-PCR utilizing cDNA derived from human epididymal RNA. Two independent reactions were performed and yielded indistinguishable results. The single molecular species amplified by each reaction was slightly smaller than the corresponding cy-ESP13.2 fragment (data not shown). It was cloned, and its nucleotide sequence was found to be 79% identical to the M. fascicularis counterpart, with two in-frame deletions relative to the macaque sequence; the largest maps to 9 residues within the Ala-Thr-rich region (Fig. 1A). We sought further corroboration of the relative deletion in hu-ESP13.2 in case it represented a Taq polymerase-induced artifact. We subjected cDNA derived from human epididymal RNA to RT-PCR using the 3' primer, #2, and a 5' primer, #3, that straddled one of the relative deletion end-points. Primer #3 can only prime synthesis if its complementary sequence is continuous within hu-ESP13.2. Primers #2 and #3 indeed produced a strong product of the expected size (data not shown), suggesting that primer #3 was functional; the deletion in hu-ESP13.2 relative to cy-ESP13.2 is thus unlikely to be artifactual.
Anti-M. fascicularis-ESP13.2 Antibodies Detect an Epididymal Protein of 30 kDa
To investigate whether the cy-ESP13.2 sequence corresponded to a protein in vivo, polyclonal antibodies were raised against a synthetic cy-ESP13.2 polypeptide (whose sequence gave no hits in database searches). The anti-peptide antibodies were used to probe Western blots containing cytosolic proteins from the testis and epididymis of mouse, rat, and macaque, and from ejaculated human seminal plasma and a 1% SDS extract of washed human spermatozoa. Results (Fig. 4) showed a strong reaction with a diffusely migrating protein of approximately 30 kDa in macaque epididymal cytosols. There was no strong reaction in the other samples. This suggests that the ESP13.2 epitope(s) recognized by the antibodies are M. fascicularis-specific and that their expression is tissue-restricted within the macaque.
|
The discrepancy between the observed size of cy-ESP13.2 (approximately 30 kDa) and that predicted by the peptide component (13.2 kDa) indicates posttranslational modification of the protein. Since ESP13.2 lacks consensus sites for N-linked glycosylation, this possibly reflects the addition of O-linked sugars on one or more of its 10 Ser or 19 Thr residues.
Immunolocalization of ESP13.2 in Macaque Epididymis and Spermatozoa
Unfixed frozen sections from the proximal and distal regions of the macaque epididymis were probed with anti-cy-ESP13.2-peptide antiserum and visualized with FITC-conjugated, goat anti-rabbit IgG (Fig. 5). A reaction was observed within the principal cells of the epididymal epithelium, including the efferent ductules and extending along the length of the duct to the cauda region (Fig. 5). The fluorescence was more intense in proximal than distal regions of the duct, and was present in both apical and basal cytoplasm (Fig. 5). However, only a weak reaction—barely above the preimmune background—was obtained in the lumen of any part of the duct.
|
To ascertain whether an association between cy-ESP13.2 and spermatozoa could be demonstrated, frozen epididymal tissue was thawed, and slices were removed from the proximal and distal regions and minced in PBS. Liberated spermatozoa were smeared onto glass slides, air-dried, and probed with anti-ESP13.2 antiserum. A specific reaction on spermatozoa was not observed, regardless of the epididymal region from which they were removed (data not shown). Likewise, freshly ejaculated macaque spermatozoa that had been washed twice in PBS and labeled in suspension, and similarly treated frozen-thawed human spermatozoa from a fertile donor, were not labeled (data not shown).
ESP13.2-Related Sequences Revealed by Database Searching
The cy-ESP13.2 sequence was used to search the EMBL database (version 56.0). No close match was found initially among annotated sequence entries. However, several expressed sequence tag (EST) entries showed significant hit scores. Three distinct human EST consensus sequences were identified after assembly of fragments with the GCG (version 9.0) program Gelassemble (Fig. 6A). Alignment with these three sequences (designated 1est, 2est, and 3est) revealed a conserved pattern of 6 cysteine residues and a putative N-terminal membrane targeting signal (Fig. 6A). Hu-ESP13.2 is identical to the est2 consensus (Fig. 6A). The C-terminal portions of the sequences were divergent with respect to length and composition. These data suggest that ESP13.2 is a member of a family of small, secretory proteins possessing multiple internal stabilizing disulfide bridges; C-terminal diversity indicates that the proteins are rapidly evolving, with differences reflecting functional specificity.
|
Further database interrogation revealed matches to extracellular molecules that contain a four-disulfide core signature made up of eight cysteine residues; ESP13.2 is predicted to conserve three of the four disulfides (Fig. 6B). These proteins represent diverse biological contexts, and include the human serine protease inhibitor, 2REL [18], the scorpion venom neurotoxin, 2SN3 [19], and the horseshoe crab peptide antibiotic, defensin [20]. The ESP13.2 core has therefore most likely been recruited from an ancient protein motif.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
We did not detect ESP13.2 or its transcripts in rat epididymal samples. One interpretation of this is that the rat lacks an ESP13.2 ortholog. Database searches with cy-ESP13.2 (or hu-ESP13.2, 2est) failed to detect strong matches. However, searches with the related sequence, 1est, identified two highly conserved ESTs from the rat (accession numbers ai044271 and ai044467) (data not shown). Identification of presumed rat orthologs of 1est argue that the ESP13.2 family of proteins performs a conserved function in mammals. One or both of the 1est-like proteins in rodents may thus fulfil the roles played by epididymal ESP13.2-like sequences in humans and other primates.
Further interrogation of databases with the ESP13.2 sequence identified a functionally and evolutionarily diverse family of proteins sharing a four-disulfide core signature, whose prototype is whey acidic protein [24]. Representatives of the family include a serine protease inhibitor [18], a scorpion venom inhibitor of Na+ channels [19], a horseshoe crab peptide antibiotic [20], and the inhibitor of sperm Ca2+ uptake, caltrin [25].
It has been proposed that caltrin-mediated inhibition of Ca2+ uptake may regulate capacitation and/or acrosomal exocytosis [25, 26]. Like primate ESP13.2, bovine caltrin-I is a basic, epididymal protein, but, unlike ESP13.2, it is small (5.5 kDa) and binds the sperm plasma membrane [27, 28]. Human ESP13.2 is thus unlikely to be orthologous to bovine caltrin-I (or the related guinea-pig proteins caltrin-I and -II; [25]). However, this does not preclude a role for ESP13.2 in regulating epididymal Ca2+ uptake and possibly the acquisition of forward motility during maturation. Such regulation would presumably involve only a weak interaction between ESP13.2 and sperm Ca2+ channels since we were unable to detect cy-ESP13.2 on spermatozoa.
Alternatively, ESP13.2 may function as a protease inhibitor. Several members of the four-disulfide core family inhibit proteases, including elafin [29] and antileukoprotease [30]. Epididymal secretions contain high levels of serine protease inhibitors whose prototype is acrosin trypsin inhibitor [31]. Sperm transit through the epididymis is accompanied by stage- and site-specific surface proteolysis as exemplified by the processing of fertilin ß [2, 7, 11]; there is evidence that fertilin ß processing involves one or more epididymal serine protease activities [32]. The tight regulation of protease(s) could be linked to the abundance of at least two putative epididymal protease inhibitors [31, 33,34]. Hence, there exists a candidate role for ESP13.2 in regulating sperm endoproteolysis.
Recently, it has been proposed that molecules synthesized by the epididymis might be transported to spermatozoa via extracellular vesicles analogous to prostasomes [12, 35]. At least one abundant epididymal protein [36] is secreted with a glycosylphosphatidylinositol (GPI) moiety attached, consistent with its presence in such extracellular membrane vesicles [8]. These vesicles would probably entrap multiple membrane proteins, including the orphan transmembrane protein, EAPI [16]. EAPI may be of note because it contains protease-like and integrin-ligand domains. Moreover, EAPI is conserved in humans (unpublished results and Fig. 3B), as well as macaques and rodents [16], further emphasizing conservation of function, plausibly in mediating epididymal vesicle-sperm docking.
Stringent regulation would be a feature of fusion between spermatozoa and prostasome-like vesicles. The epididymal abundance of molecules with the potential to inhibit proteases suggests a mechanism of endoproteolytic unmasking of one or both partners in a sperm receptor/vesicle ligand pair. This is a function for ESP13.2 consistent with its secretory nature. Moreover, ESP13.2 is not the sole four-disulfide core family member expressed in the epididymis [25, 33]. Given this and the high level of ESP13.2 expression, it is likely that the ESP13.2 family of four-disulfide core proteins plays an important role in primate epididymal functionality and possibly sperm maturation.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
2 Correspondence. FAX: 1 808 956 5474; perry{at}hawaii.edu
Accepted: May 13, 1999.
Received: April 8, 1999.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  T. L. Tollner, A. I. Yudin, A. F. Tarantal, C. A. Treece, J. W. Overstreet, and G. N. Cherr Beta-Defensin 126 on the Surface of Macaque Sperm Mediates Attachment of Sperm to Oviductal Epithelia Biol Reprod, March 1, 2008; 78(3): 400 - 412. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. Dube, P. T.K. Chan, L. Hermo, and D. G. Cyr Gene Expression Profiling and Its Relevance to the Blood-Epididymal Barrier in the Human Epididymis Biol Reprod, June 1, 2007; 76(6): 1034 - 1044. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Zanich, J. C. Pascall, and R. Jones Secreted Epididymal Glycoprotein 2D6 That Binds to the Sperm's Plasma Membrane Is a Member of the {beta}-Defensin Superfamily of Pore-Forming Glycopeptides Biol Reprod, December 1, 2003; 69(6): 1831 - 1842. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. I. Yudin, T. L. Tollner, M.-W. Li, C. A. Treece, J. W. Overstreet, and G. N. Cherr ESP13.2, a Member of the {beta}-Defensin Family, Is a Macaque Sperm Surface-Coating Protein Involved in the Capacitation Process Biol Reprod, October 1, 2003; 69(4): 1118 - 1128. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Rao, J. C. Herr, P. P. Reddi, M. J. Wolkowicz, L. A. Bush, N. E. Sherman, M. Black, and C. J. Flickinger Cloning and Characterization of a Novel Sperm-Associated Isoantigen (E-3) with Defensin- and Lectin-Like Motifs Expressed in Rat Epididymis Biol Reprod, January 1, 2003; 68(1): 290 - 301. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. H. Hall, K. G. Hamil, and F. S. French Host Defense Proteins of the Male Reproductive Tract J Androl, September 1, 2002; 23(5): 585 - 597. [Full Text] [PDF]
|
|
|
|
 |
|
 C. Kirchhoff The dog as a model to study human epididymal function at a molecular level Mol. Hum. Reprod., August 1, 2002; 8(8): 695 - 701. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Q. Liu, K. G. Hamil, P. Sivashanmugam, G. Grossman, R. Soundararajan, A. J. Rao, R. T. Richardson, Y.-L. Zhang, M. G. O'Rand, P. Petrusz, et al. Primate Epididymis-Specific Proteins: Characterization of ESC42, a Novel Protein Containing a Trefoil-Like Motif in Monkey and Human Endocrinology, October 1, 2001; 142(10): 4529 - 4539. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
