| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Behavior |
Department of Fisheries and Wildlife,3 Michigan State University, East Lansing, Michigan 48824
The Center for Environment, Fisheries and Aquaculture Science,4 The Nothe, Weymouth, Dorset DT4 8UB, United Kingdom
Department of Biological Sciences,5 University of Windsor, Windsor, Ontario N9B 3P4, Canada
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, 12, 24-trihydroxy-5-cholan-3-one-24-sulfate (3 keto-petromyzonol sulfate, 3ketoPZS) that induces search and preference behaviors in ovulating females. In this study, we conducted a series of experiments to demonstrate that release of this pheromone into water takes place exclusively through the gills. In a behavioral maze, water conditioned with the anterior region of spermiating males induced an increase of search and preference behaviors in ovulating females. Similar behavior was not elicited by water conditioned by the posterior region. The anterior region washings and whole-body washings from spermiating males also elicited large and virtually identical electro-olfactogram responses from female sea lampreys, while the posterior washings produced negligible responses. Further, mass spectrometry and immunoassay confirmed that virtually all the 3ketoPZS released into water was through the gills. Immunocytochemistry revealed some gill epithelial cells and hepatocytes from spermiating males contained dense immunoreactive 3ketoPZS, but not those from prespermiating males. These results demonstrate that 3ketoPZS is released through the gill epithelia and suggest that this pheromone or its precursor may be produced in the liver.
environment, male sexual function, pheromones, spermatogenesis
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
,12,24-trihydroxy-5-cholan-3-one-24-sulfate (3 keto-petromyzonol sulfate, 3ketoPZS), which acts as an attractant for ovulating females [1]. Although this sex pheromone has been identified and its function demonstrated, its mode of synthesis and excretion remains elusive. Bile acids are typically produced in the liver, secreted into the gallbladder, and excreted through the intestine along with feces [2, 3]. Larval sea lampreys appear to have evolved this route as well [4, 5]. However, this same route is not available to adult sea lampreys, which lack gallbladders and bile ducts [6]. In river lamprey Lampetra fluviatis, the gills of spawning males contain large glandular cells [7, 8] that have been postulated to "secrete some substance of sexual significance" [7]. More recently, we showed that water from the anterior region of spermiating males (bathing the gills) contained far more immunoreactive 3ketoPZS than water from the posterior region [9] and suggested that gills may mediate the release of this pheromone compound [1]. However, there was no proof that this compound emanated from the gills rather than from some other part of the anterior region and certainly no direct link to the glandular cells. Further, there was no corroborating chemical and biological evidence from any other procedure that 3ketoPZS was indeed emanated mainly from the anterior region.
The main hypothesis that we address in the present study is that the male pheromone is released into the water via the glandular cells of the gills. Our expectations are 1) that the results of ELISA [9], which showed that 3ketoPZS was released largely from the anterior region of spermiating males, can be confirmed by mass spectrometry, electro-olfactogram, and behavior induction in ovulating females; 2) that water emitted by gills (before it touches any other tissues) contain 3ketoPZS at a level comparable to that estimated from anterior body washings; 3) that the unusual glandular cells in the gills will stain immunocytochemically with antibodies to 3ketoPZS; and 4) that the gills of spermiating males have many more of these cells than the gills of prespermiating males.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This research was approved by the Michigan State University, All University Committee on Animal Use and Care, and complied with all federal and state laws, policies, and rules for the humane use of laboratory animals in research. Adult sea lampreys were collected from tributaries to Lakes Huron and Michigan by the staff of the U.S. Fish and Wildlife Service, Marquette Biological Station, Marquette, MI. The animals were transported to the main laboratory at the U.S. Geological Survey, Hammond Bay Biological Station, Millersburg, MI. Males and females were separated and held in flow-through tanks (1000 L) with Lake Huron water at temperatures ranging from 7°C to 20°C. Males and females were checked periodically for spermiation and ovulation according to the criteria and procedures reported [10]. Any spermiating and ovulating individuals were separated into two other tanks, respectively.
Collection of Washings and Extracts
Washings from the anterior and posterior regions of spermiating males were collected using a bisected, acrylic aquarium (Fig. 1). A divider with a hole to accommodate a sea lamprey's head was fixed in the middle of the aquarium to make two separate chambers. The hole was lined with a latex gasket that, when the male was in place, prevented water from flowing between the two chambers. A perforated acrylic tube was mounted on one side of the hole to immobilize the anterior region. The posterior region was held in flexible plastic mesh that was tightened according to the size of the male.
|
During a washing, a spermiating male was anesthetized with tricaine methane sulfonate (MS-222; Argent Chemical Laboratories, Redmond, WA) and placed headfirst through the gasket/divider, making sure all the gills were located in the forward chamber. The plastic mesh tube around the posterior region was then tightened. The anterior chamber was filled with 7 L of water, the latex gasket inspected for leaks, and then the posterior chamber filled with 7 L of water. Each chamber was aerated, and the male was held this way for 1 h. At the end, all the water was siphoned into buckets and either used directly in experiments or extracted as previously reported [1]. For extractions, water was prefiltered with No. 3 Whatman filter paper (Millipore Corp., Bedford, MA) and then drawn through activated Sep-Pak octadecylsilane cartridges (Waters, Milford, MA). These were then washed with distilled water, eluted with methanol, and stored at -80°C.
To prove that there was no leakage across the latex gasket in the bisected aquarium, two experiments were performed both before and after the washing collection period. In the first, males were secured in the aquarium. A dye was then added to one chamber and the other checked visually after 1 h (Fig. 1B). The aquarium was then drained and the experiment repeated, adding dye to the other chamber, checking for leakage in the other direction. In the second experiment, females, known not to release 3ketoPZS [1, 9], were placed in the aquarium, and 1 mg of synthetic 3ketoPZS was introduced into one of the chambers. After 1 h, water samples were taken from both chambers and analyzed for 3ketoPZS by ELISA [9]. Again, the aquarium was drained and the experiment repeated, adding synthetic 3ketoPZS to the other side, checking for leakage in the other direction.
Behavioral Assays on Water Taken from the Anterior and Posterior Regions of Spermiating Males
The sex pheromone released by spermiating males induces preference and search behaviors from ovulating females [1, 10]. Therefore, we measured these two types of characteristic behavioral responses from ovulating females to chemical stimuli from the anterior and posterior regions of spermiating males using the identical apparatus and protocols developed in the previous studies [1, 10]. Briefly, preference behavior was the amount of time spent in either side of a two-choice maze, while search behavior was the amount of time spent swimming at the head of either side of a two-choice maze. In each test, the behavior of a single female was video recorded before and after the introduction of washings from either the anterior or the posterior region of spermiating males into the odor chamber of the side of the maze chosen randomly by the toss of a coin. Because the same females were used to test both washings, the order of washing presentation was also randomized. The washings were delivered at 75 ml/min using a peristaltic pump. Tests were conducted between 0700 and 1700 in water temperatures that ranged from 12°C to 24°C. Preference and search behaviors were scored by naive observers and the data analyzed with a two-tailed Wilcoxon signed rank test [1, 10, 11].
Electro-Olfactographic Recordings on Water from the Anterior and Posterior Regions of Spermiating Males
Synthetic and natural 3ketoPZS induce strong electro-olfactographic (EOG) responses from adult female sea lampreys [1], as does water conditioned by spermiating male sea lampreys [10, 12]. Using EOG recording on females, we determined the olfactory potency of washings from the anterior and posterior regions and the whole body of spermiating males and established concentration-response relationships, according to established procedures [10, 13]. For each recording, a 10-5-M L-arginine standard was pulsed into the olfactory epithelium of a female and the EOG response measured to establish a baseline of electrical activity. Next, blank control water was introduced and the response measured to confirm the absence of odorants from the clean water source used to perfuse the olfactory epithelium (and also to mix the test odorants). Increasing concentrations of test odorants (starting at 106 dilution) were then introduced and the responses measured. Measuring the response to the L-arginine standard and blank controls concluded each trial. The epithelium of each female was allowed to recover for at least 3 min between stimuli, and each concentration of an odorant was tested at least twice on a female. The magnitude of EOG responses was measured [14] and expressed as a percentage of the L-arginine standard [13].
Mass Spectrometry Analysis of Water and Urine from Spermiating Males
Methanol extracts of washings (1 L) taken from the anterior and posterior regions of spermiating males were dried down under a stream of oxygen-free nitrogen gas at 45°C, reconstituted in chloroform, and subjected to fast atom bombardment mass spectrometry (FAB MS, 10KV) in both negative and positive ionization modes.
To determine whether 3ketoPZS was present in the urine, spermiating males were anesthetized with an MS-222 solution, and a catheter was inserted into the urogenital pore [15]. Urine was collected into a 50-ml container that was emptied periodically and stored at -80°C until analysis. The urine was passed through a Sep-Pak and the eluate dried down and subjected to FAB MS as described previously.
ELISA of 3ketoPZS
The procedure for ELISA of 3ketoPZS has been described [9]. The antibody shows 100% cross-reaction with 7,12,24-trihydroxy-5-cholan-3-one (3 keto-petromyzonol, 3kPZ) and 7,12-trihydroxy-5-cholan-3-one-24-oic acid (3 keto-allocholic acid, 3kACA). However, the former was not found in washing extracts from spermiating males [9], and the latter was present at a ratio of only 1:25 (3kACA:3ketoPZS [16]). The anterior and posterior washings from spermiating males were diluted 1:25 with assay buffer before ELISA.
ELISA was also carried out on water collected directly from the gills of spermiating males. To do this, a male was anesthetized with metomidate hydrochloride (Syndel, Vancouver, BC, Canada), immobilized with gallamine triethiodide (Sigma, St. Louis, MO), and placed in a flow-through trough. Water flowed through the mouth and exited the gills at an average rate of 431 ml/min. Water was pipetted directly off the gills and placed into a beaker every 3 min for 0.5 h, making a total of 10 individual samples for each spermiating male. Parts of each sample were stored at -80°C, and 100 ml of each sample were extracted using a Sep-Pak, and the eluate was also stored at -80°C until ELISA analysis.
Hourly release rates of 3ketoPZS were calculated for the anterior region washings and water collected directly from the gills. For the anterior washings, the 3ketoPZS level of water was multiplied by the total volume used in the anterior portion of the aquarium (7 L). For water collected directly from the gills, the average 3ketoPZS level of water for each spermiating male was multiplied by the flow of water across the gills (431 ml/min) and then by 60 min.
Immunocytochemistry for 3ketoPZS in the Gills and Liver of Male Sea Lampreys
Five spermiating and seven prespermiating males were killed with an overdose of MS-222. The gills and liver were removed, immersed in either Zamboni's fixative (2% paraformaldehyde and 1.2% saturated picric acid in 0.1 M phosphate buffer [PB]) or 4% paraformaldehyde (in PB) and stored at 4°C. As the male pheromone is a steroidal lipid, whole gill filaments were mounted to minimize the loss of cellular lipids [17]. Each gill pouch was placed under a dissection microscope, and single gill filaments were removed. Throughout the immunostaining procedure, the gill filaments were agitated mildly at 4°C. The single gill filaments were rinsed three times over 1 h with either PB or with PB plus 0.1% Triton X-100 (PB-TX) for 30 min. The omission of Triton-X from the immunostaining protocol served to assist in keeping the cellular membranes and lipid-containing subcellular structures intact. Nonspecific binding was blocked with 5% normal goat serum in PB for 20 min, drained off, and exchanged for primary antibody (from four rabbits: codes 184, 185, 286, 285 [9]) diluted with 0.1 M PB or 0.1 M PB-TX (1:1000–1:5000) and incubated for 24 h. The tissue was then rinsed three times over 1 h with 0.1 M PB, followed by a 3-h incubation in Alexa 488 goat anti-rabbit IgC, which excites at 488 nm (1:100; Molecular Probes, Eugene, OR); rinsed; and mounted in glycerol or viewed directly with a 40x water immersion lens. Following immunostaining, some filaments were sectioned with a vibratome (Leica Microsystems, Wetzlarm, Germany) to allow for viewing of the interplatelet region. The immunostained filament was embedded in 5% agarose (American Bioanalytical, Natick, MA) and sectioned at a thickness of 150 µm. All preparations were viewed using a BioRad 1024 Confocal Microscope (Bio-Rad, Hercules, CA) with the following settings: laser power, 10%; iris, 1.5 mm; gain, 1000x; black level, 0. Fixed liver samples were embedded in 5% agarose, sectioned using a vibratome at a thickness of 200 µm, and stained for 3ketoPZS by the procedure used for the gill whole mounts (see previous discussion). Liver samples were viewed with a 60x oil immersion lens using the same settings.
A preadsorption control experiment was conducted using a 10-fold molar excess, relative to IgC concentration, of antigen (3ketoPZS) to antibody dilution. The 3ketoPZS/antibody mixture was incubated for 24 h at 4°C, with mild agitation, then centrifuged at 100 000 g for 30 min (Sorvall RC M120 GX, Kendro, Asheville, NC). The supernatant was used in place of the primary antibody in the previously described immunocytochemical protocol. The preadsorption control did not contain any immunostaining. Negative controls were also conducted using preimmune sera for antisera from rabbits 184 and 185 [16]. Tissues processed with these preimmune sera were unstained. In addition, a control with the primary antibody omitted was included in each experiment.
Electron Microscopic Analysis of Gill Tissues
The gill tissues from spermiating and prespermiating male sea lampreys were fixed in buffered 2.5% glutaraldehyde and then in 1% OsO4. After dehydration, fixed tissues were embedded in Spurtol resin. Ultrathin sections (90 nm) were cut and then stained with uranyl acetate and lead citrate. The sections were examined with a Philips CM10 electron microscope.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
In EOG experiments, the mean response to 10-5 M L-arginine was 0.5 mV (SE = 0.09 mV; n = 6). The anterior region washings (n = 6) of spermiating males were far more potent than those of the posterior region (n = 5) at equivalent dilutions (Fig. 2) but were equipotent to those from whole spermiating males (n = 6). The detection threshold for the posterior region washings was approximately 1:100 (v/v), whereas that for the anterior region and whole-body chemical stimuli was approximately 1:10 000 (v/v).
|
Negative FAB MS analyses showed that the most abundant ion in extracts of water from the anterior region of spermiating males was at m/z 471 (n = 6; Fig. 3A), the same as synthetic 3ketoPZS [1]. At positive mode, the base peak was at m/z 473. FAB MS detected only trace amounts of this molecule in the extracts of water from the posterior regions (n = 6; Fig. 3B) and in extracts of spermiating male urine (n = 4; data not shown).
|
The mean concentration of immunoreactive 3ketoPZS in water collected from the anterior regions of spermiating males was 54 ng/ml (n = 13), whereas that from the posterior regions was 0.8 ng/ml (n = 13). On average, the total release rate of the anterior region was 378 µg/h (Fig. 4). Also, immunoreactive 3ketoPZS was found in water collected directly off the gills of five spermiating males; however, these concentrations showed considerable variation (Fig. 4). On average, the release rate of 3ketoPZS estimated from water collected directly off the gills was 308 µg/h. There was no significant difference between release rates estimated from the anterior region water and gill water (Student t-test, P > 0.10).
|
The gill filaments were comprised of gill platelets (lamellae; Fig. 5A) covered by low cuboidal platelet cells, with cuboidal to columnar shaped cells in interplatelet regions. In prespermiating males, the pheromone immunoreactivity was weak and diffuse on the platelet surface and granular in interplatelet epithelial cells (antibody 286; Fig. 5, A–C). In gill tissue from spermiating males, subpopulations of platelet cells were strongly immunoreactive to the pheromone antibodies (antibody 185; Fig. 5, D and E). This staining was absent from the nucleus (Figs. 5E and 6A). Both platelet and interplatelet epithelial cells contained supranuclear regions that were intensely immunoreactive to the pheromone antibodies (antibody 286; Fig. 6, A–C). These immunopositive cells were cuboidal along the platelet base regions (Fig. 6A) and columnar in the interplatelet regions (Fig. 6C). In all samples, pheromone immunoreactivity was stronger and clearer when Triton-X was omitted from immunostaining procedure than in samples treated with Triton-X (antibody 185; Fig. 6, D and E). The fact that the immunoreactivity decreased and blurred following treatment with the lipid solubilizing detergent is consistent with the lipid nature of the pheromone.
|
|
Transmission electron microscopy of the gill epithelial cells revealed low microvillar cuboidal platelet cells (Fig. 7A) with small electron lucent vesicles, larger electron opaque granules (0.3 µm), round mitochondria, rough and smooth endoplasmic reticulum, and Golgi apparatus. In the basal platelet region and the interplatelet region, the cuboidal cells (Fig. 7B) had a dome-shaped apical surface, smooth endoplasmic reticulum, rough endoplasmic reticulum, electron dense vesicles, dense core vesicles, and electron lucent vesicles. The supranuclear region of low columnar interplatelet cells (Fig. 7C) contained electron opaque granules and smooth endoplasmic reticulum. The perinuclear cytoplasm of interplatelet cells (Fig. 7D) contained granules of varying sizes and electron density and numerous vesicular and tubular cisternae of smooth endoplasmic reticulum. These ultrastructural characteristics show that these cells are active in lipid metabolism and store secretory products.
|
In spermiating male liver, the immunostaining (antibody 285) was strong and diffuse in hepatocyte cytoplasm and strong in widespread cytoplasmic granules (Fig. 8A). The liver of prespermiating males displayed pheromone immunoreacitivity (antibody 285) that was weak and diffuse, with scattered intensely stained cytoplasmic granules (Fig. 8B). Therefore, the pheromone is present in higher levels in the spermiating male liver than in the liver of the prespermiating male.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The EOG and ELISA data also indicate that the 3ketoPZS released from the gills accounts for all the 3ketoPZS released by the whole animal, supporting the second subhypothesis. The anterior region washings induce EOG responses at virtually the same magnitude as the whole washings did over a range of dilutions that spans four orders of magnitude. The average release rate from the anterior region, 378 µg/h, is close to the previously reported whole-body release rate of approximately 500 µg/h [9]. Most important, the washings collected directly from the gills contain large amounts of 3ketoPZS (
308 µg/h), which can account for the majority of the 3ketoPZS release estimated from the washings collected from the anterior region. These results support the gill-release hypothesis set forth in our previous study [1].
Finally, our third subhypothesis is clearly supported by immunocytochemical experiments that show the presence of 3ketoPZS in cells of both the liver and the gills. It appears likely that the pheromone, a sulfate- and ketone- containing bile acid derivative, moves into the water from the platelet and interplatelet cells of spermiating males. Diffuse immunolabeling on the surface of platelet cells of both stages may be due to the secreted pheromone adhering to the cell surface of these cells. The localization of pheromone immunoreactive granules in the interplatelet cells of prespermiating males suggests that pheromone-containing cells in the platelet epithelium first appear in the interplatelet region during the prespermiating phase and then only in the platelet region during spermiation. It is likely that 3ketoPZS is at least a part of the "substance of sexual significance" [7].
Our results clearly implicate the involvement, but do not illustrate the explicit role, of hepatocytes in biosynthesis of 3ketoPZS. The granules with strong pheromone immunoreactivity demonstrate that hepatocytes contain either 3ketoPZS, 3kACA, 3kPZ, or all of them because the primary antibody used for immunocytochemical staining cross-reacts with all three of these compounds. This conforms with the discovery that PZS, a larval bile acid, is produced in the liver [5] and that liver is the exclusive organ for bile acid synthesis [18]. It is possible that 3ketoPZS is synthesized in the liver, released into circulation, and taken up by the platelet and interplatelet cells. Alternatively, the gill cells may take up a precursor synthesized in the liver and modify it into 3ketoPZS. Smooth endoplasmic reticulum, a subcellular site for steroid synthesis, is widespread in the platelet and interplatelet cells. Granules of varying electron density that are prominent in transmission electron micrographs may be the site of the granular localization of pheromone immunoreacitivity of the interplatelet cells. Large lipid structures with electron lucent centers, previously observed in Type I male glandular cells of river lamprey fixed in osmium tetroxide [8], were absent from sea lamprey. This difference may be due to the differing fixation protocols, with stabilizing aldehyde prefixation being included in the present study but absent from the earlier study.
In bony fishes, all pheromones identified to date are steroids or prostaglandins—or sulfated or glucuronidated forms of these compounds [19–22]. In the case of free steroids, there is evidence that they, like the lamprey pheromone, are mainly excreted into the water via the gills [23, 24]. However, the mechanism whereby they do so appears to be passive diffusion [23, 25]. There is no evidence for specialized cells such as those that we have demonstrated in lampreys. The release rates of free steroids by bony fishes are also considerably lower than those of 3ketoPZS by spermiating lampreys [1, 26]. In contrast to free steroids, sulfated steroids in teleosts and elasmobranches do not pass through the gills [23, 25]. Their main site of release appears to be the urinary bladder [15, 24]. Although the urinary bladder has been thought as a source of sex pheromones in the sea lamprey [27], none has yet been firmly identified. The main reason for making these comparisons between vertebrate classes is to underline our belief the mechanism whereby male sea lampreys release 3ketoPZS is an active one, with the specialized cells that appear in the gill epithelium of spermiating males as "pumps."
To rely on gills to broadcast 3ketoPZS possibly represents the evolution of a system mainly to extend the active space of the pheromone signal. Such a system would certainly be advantageous for spermiating male lampreys, which construct nests in sites where water flows at 0.5 to 1.5 m/sec [28, 29] and probably have to rely heavily on this pheromone to attract ovulating females from downstream [1, 27].
Bile acids and their sulfate esters, such as 3ketoPZS, are produced in large quantities by hepatocytes of vertebrates [18]. The transportation of 3ketoPZS from liver to gills is potentially very efficient because, in the lamprey, hepatic veins carry blood directly to the heart, and all blood from the heart passes immediately through the gills [30]. It has been estimated that in a 100-g river lamprey, the relative area of the gills is 600 mm2/g [31]. If the relative surface area in the sea lamprey were similar, its gills would provide an enormous surface area with specialized glandular cells, through which the respiratory activity generates continuous flow to facilitate exchange of the pheromone molecule with the environment.
In conclusion, our experiments demonstrate that the spermiating male sea lamprey sex pheromone, 3ketoPZS, is released through the gills. Although liver and glandular cells are clearly implicated in the production and release of 3ketoPZS, their explicit roles need to be examined at the molecular level to elucidate the mechanistic processes whereby this pheromone molecule is produced and released.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
2 Correspondence: Weiming Li, Department of Fisheries and Wildlife, Michigan State University, 13 Natural Resources Building, East Lansing, MI 48824. FAX: 517 432 1699; liweim{at}msu.edu
Received: 10 December 2002.
First decision: 30 December 2002.
Accepted: 11 February 2003.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
,20-dihydroxy 4-pregnen-3-one 20-sulfate, and of the glucuronides of testosterone and 17ß-estradiol, by urine of reproductively mature male and female rainbow trout (Oncorhynchus mykiss). J Fish Biol 1994 44:117-129-dihydroxy-4-pregnen-3-one. Gen Comp Endocrinol 1996 101:180-194[CrossRef][Medline]This article has been cited by other articles:
|
|
 |
|
  W. Li Potential Multiple Functions of a Male Sea Lamprey Pheromone Chem Senses, January 1, 2005; 30(suppl_1): i307 - i308. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
