| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Hopkins Marine Station and Department of Biological Sciences of Stanford University, Pacific Grove, California 93950
2 Department of Chemistry and the Beckman Institute, University of Illinois, Urbana, Illinois 61801
3 Department of Pharmacology, Yale University School of Medicine, New Haven, Connecticut 06520
* To whom correspondence should be addressed. E-mail: lignje{at}stanford.edu
 |
Abstract |
|---|
 TOP Abstract IntroductionMaterials and MethodsResultsDiscussionLiterature Cited |
|---|
Abbreviations: MALDI-TOF, matrix-assisted laser desorption-ionization time of flight mass spectrometry • m/z, mass-to-charge ratio
|
Introduction |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionLiterature Cited |
|---|
Each of the more than 500 Conus species (Röckel et al., 1995; Kohn, 1998) is thought to contain a large number of distinct peptide toxins in the venom duct (Olivera, 1997; McIntosh et al., 1999), and contemporary Conus research has primarily focused on purifying and characterizing these toxins. Conus toxins are typically 10–30 amino acids in length and belong to one of several families defined by highly conserved cysteine frameworks and internal disulfide linkages (Olivera et al., 1995; McIntosh et al., 1999; Duda and Palumbi, 1999; Conticello et al., 2001). Most of these peptides target voltage- and ligand-gated ion channels.
Production and delivery of Conus venom involves three general steps: (1) synthesis, processing, and packaging of peptide toxins; (2) generation and storage of radular teeth and transfer of a tooth to the tip of the proboscis; and (3) the final insertion of the tooth and ejection of venom. These processes are carried out within an anatomically complex venom apparatus in conjunction with the anterior regions of the digestive tract. These features are relatively invariable between Conus species (see Halstead, 1988), and a schematic venom apparatus is illustrated in Figure 1.
|
This report describes some new features of the venom apparatus in C. californicus as revealed by conventional light and transmission electron microscopy. First, we have found that the proximal portion of the venom duct has a unique structure and may be specialized for active transport rather than secretion. Second, we have discovered a novel and unusually narrow passageway between the venom duct and the pharynx through which venom must pass prior to ejection. Third, complex, granular material resembling that in the lumen of the venom duct is found packaged inside the lumen of individual teeth while they are still in the radular sac. Furthermore, mass spectrometry demonstrates that presumptive peptides can be identified in material isolated from the venom duct and from the lumen of single radular teeth. Although the sequences of these peptides are unknown, equivalence in their apparent molecular weights suggests that some of the same peptides are present in both locations.
|
Materials and Methods |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionLiterature Cited |
|---|
Light and electron microscopy
In preparation for light microscopy, two animals were relaxed in seawater containing 2% MgCl2, and the shells were cracked with vise-grips and removed. The complete venom apparatus and anterior digestive tract as depicted in Figure 1 were dissected under seawater. Tissue samples were fixed overnight at 4°C in 2% glutaraldehyde made up in 80% filtered seawater containing 100 mM HEPES (pH 7.3) and then dehydrated in a graded ethanol series. Tissues were then cut into smaller pieces (several cubic millimeters) and embedded in Spurs resin. Sections (0.5–2 µm) were cut with an ultramicrotome equipped with a glass knife, stained with 1% methylene blue dissolved in 0.1 N NaOH, and examined with an Olympus BH-2 microscope. The venom duct was examined at several places along its length, and anatomical locations and orientations of all sections figured in this paper are indicated in Figure 1.
For transmission electron microscopy (TEM), tissue was fixed and embedded as described above, except a post-fixation step of 1% OsO4 in seawater for 1 h was added. Sections were cut with a diamond knife, stained with saturated uranyl acetate and lead citrate in 0.1 N NaOH, and viewed on a Phillips CM12 microscope at Stanford University School of Medicine.
Living animals were also relaxed and dissected as described above to reveal the junction between the short arm of the radular sac and the pharynx. An incision was made at the junction and several teeth were carefully removed with a fine forceps. The teeth were rinsed well in seawater and examined using water immersion optics.
Matrix-assisted laser desorption-ionization time of flight mass spectrometry (MALDI-TOF)
A complete venom apparatus was removed as described above, and the distalmost thin segment of venom duct (area 4 in Fig. 1) was ligated with fine surgical silk close to where it enters the pharyngeal wall; this ligation prevents any leakage of venom into the proboscis and potentially into the radular sac. Most of the proboscis and esophagus were removed, and the pharynx was then slit longitudinally and the lumen was washed with seawater. This semi-intact venom duct (with bulb) and radular sac preparation was then shipped in iced seawater to the University of Illinois by overnight carrier.
Individual teeth from the short arm of the radular sac were dissected upon arrival in Illinois and washed well in isotonic NaCl. Single teeth were broken open with forceps and mashed onto a MALDI target into 0.5 µl of matrix solution (10 mg ml-1 2,5-dihydroxybenzoic acid (ICN Pharmaceuticals, Costa Mesa, CA)) in deionized water. Short (
1 mm) segments of the venom duct were also cut from locations corresponding to areas 2, 3, and 4 in Figure 1. Some of the luminal contents of these sections was squeezed into the same volume of matrix solution on additional target spots.
MALDI-TOF MS analysis was carried out as previously described (Sweedler et al., 2000; Li et al., 2000) using a Voyager DE STR (PE Biosystems, Framingham, MA). A pulsed, nitrogen laser (337 nm) served as the desorption/ionization source, and positive-ion mass spectra were obtained with both linear and reflection modes of operation. The former mode has greater sensitivity, whereas the latter is more accurate in establishing mass. Mass-to-charge ratio (m/z) calibrations were performed externally with a standard calibration mixture including angiotensin and bovine insulin (Sequazyme, PE Biosystems, Framingham, MA). For data presented in this paper, the m/z parameter is equivalent to molecular weight (Bingham et al., 1996). The spectra shown are representative examples from several hundred spectra acquired from several snails.
|
Results |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionLiterature Cited |
|---|
|
The inner surface of this collagenous layer is an amorphous basal lamina (see also Fig. 3B) that covers the cellular layer (ep). Cells in this epithelium contain numerous mitochondria (mt, white pointers) that are concentrated on the basal surface facing the collagen layer. Nuclei also tend to be located here (n in Fig. 2A). The apical surface of the epithelial cells is densely covered with microvilli that protrude into the lumen (mv in Fig. 2B). Boundaries between these interdigitating cells are highly complex in this region, and tight junctions are also present along the apical surface (black arrow).
|
The distal venom duct
About one-quarter of the distance to the pharynx, the venom duct enlarges to a diameter of about 375 µm and becomes creamy white. Duct morphology also dramatically changes, as evidenced by the light micrograph in Figure 3A. The outer layer is still collagenous (c), but the nature of the epithelium is quite different. Large, vaguely columnar epithelial cells with poorly demarcated luminal and lateral boundaries (ep in Fig. 3A) are present, and these cells are filled with deeply stained (blue) venom granules (g) as large as 5 µm or more in diameter.
Electron microscopy confirms that the mitochondrion-rich cells with apical microvilli as described above are absent from this more distal region. Figure 3B shows the external border of the duct where the basal lamina (bl) separates the cellular and collagenous (c) layers that again run in at least two directions. Immediately beneath the basal lamina (bl, Fig. 3B), cross-sections of presumptive neuronal processes (nv) are sometimes found. Many of these processes are filled with electron-dense vesicles, which are much smaller than the venom granules (g) and are not resolvable by light microscopy.
Epithelial cells just inside the external surface of the duct in this region contain numerous intracellular venom granules. A nucleus (n) and several granules can be seen within the same membrane-limited cellular unit in Figure 3C. Granules in these more peripheral epithelial cells tend to be covered with a membranous coating (small arrows). Closer to the lumen of the duct, cellular boundaries become difficult to define, even by TEM, probably because the cells are breaking down in a holocrine-type secretory process (see Halstead, 1988).
Extracellular granules well within the central duct lumen in this region are substantially more abundant than in the proximal region described above. Analysis of 50 x 50 µm2 areas in light micrographs at the magnification of Figure 3A yielded granule counts of 58.3 ± 48.3 (distal) versus 13 ± 4 (proximal; mean ± SD, n = 3 in both cases). Density in the distal lumen was quite variable, and the area in the upper left of Figure 3A, which was included in the analysis, is at the low end of the range (25–84).
Within this deeper region of the distal venom duct, individual venom granules display distinct ultrastructure and can be separated into at least two classes (Fig. 3D). The first (1 in Fig. 3D) has an electron-dense core surrounded by a more intensely stained ring, which in turn is covered by a diffuse coating of lightly stained material. There is no sign of the membranous coating. The second class (2) has a core that stains much more lightly, but displays a similar coating of the ring and external diffuse layer. The relationship of the two granule classes to each other or to venom production in general is not obvious from the anatomy, but presumably granules without a membranous coat derive from the clearly intracellular granules described above.
Profiles of cilia are also occasionally found in the distal lumen. In living material examined with water-immersion optics, clumps of motile cilia can be seen to create microscopic vortices but not directional fluid flow. Neither the origin nor functional significance of these processes is known.
The anteriormost venom duct and the connection to the pharynx
In general, the above description of the distal duct region applies up to its anterior end, just before it joins the pharynx, where the duct diameter is about 475 µm. Figure 4A shows an oblique, but generally longitudinal, section through this region. Some of the columnar-type epithelial cells contain intracellular granules similar to those found in the lumen of the duct (arrows), but the relative cellular versus luminal distribution of granules is opposite to that in Figure 3A. Here the lumen of the duct is densely packed with granules (84.7 ± 2.1 per 2500 µm-2; mean ± SD, n = 3), and intracellular granules are sparse. Epithelial cells adjacent to the pharynx (p) appear to contain no granules at all.
|
At a distance of about 250 µm into the pharyngeal wall, the narrow venom duct passageway branches into two smaller channels, each about 10 µm wide (arrows in Fig. 4E, F). Just before the branch point, blue-stained granules are largely replaced by smaller, clear granules that continue into the finer channels. These clear granules appear to line the channel, with a few larger, stained profiles persisting in the core (single arrowheads in Fig. 4E, F). The latter objects lack the smooth profiles of venom granules, and they are stained more intensely purple like the metachromatic cells. These features suggest that the stained granular material in the core of these fine channels is distinct from either of the two types of venom granules described above, but any relationship to the metachromatic cells is uncertain. In some cases, only clear granules are evident, although some amorphous material, which appears to be matted cilia, persists in the channel core (double arrowhead in Fig. 4F).
After the duct’s bifurcation, each of the branches continues for about another 500 µm before opening into the lumen of the pharynx. Two such channels are visible in Figure 5A. The proximal ends are marked by ovals (the branch point is not visible in Fig. 5A), and the distal ends (stars) open into a common space connected to the pharyngeal lumen (1). Purple cells continue to line the channel epithelium to a point about halfway from the bifurcation to the distal opening. Beyond this point (bold arrows), the clear granules in the lumen of the channels become infrequent, and the passageways, which here are ciliated, grow a bit in width. The cuboidal epithelium forming the channels in this region is contiguous with the highly folded, ciliated epithelium of the pharynx. Thus, two pharyngeal folds extend particularly deeply and merge with the fine venom duct channels. It is at this junction that the novel purple cells lining the final venom duct passageway give way to the pharyngeal epithelium. A schematic summary of the nature of the venom duct as it crosses the pharyngeal wall is given in Figure 5B.
|
|
At the site where the teeth enter the channel passing into the pharynx, they all point anteriorly (Fig. 6E) and are enveloped by muscle of the pharyngeal wall (m). At the connection between the radular sac and the pharynx, the two structures are in intimate contact, and folds of the pharyngeal epithelium wrap around the sac (p in Fig. 6E; arrow in Fig. 6F). The packing of the basal spurs evident in Fig. 6E suggests that the teeth enter this connecting channel in a tight queue. The lumen of every tooth in this region is densely packed with a rich mixture of granules and other noncellular material (Fig. 6F).
Teeth taken from this final portion of unfixed radular sacs were also studied. Examination of such teeth under water-immersion optics confirms the presence of a large amount of granular material in the lumen (Fig. 7). This granular content does not leak out to any visually noticeable extent during several hours of observation, although FITC-dextran (fluorescein isothiocyanate-dextran, 19,000 mw) can enter the lumen and after an exchange of bathing solution can be visualized by fluorescence microscopy.
|
|
|
Discussion |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionLiterature Cited |
|---|
Along with the general anatomical similarities, we have identified several features in C. californicus that have not been described in other species. Whether these features are characteristic of other Conus species remains to be established:
The proximal venom duct
Previous studies have indicated that material extracted from the proximal portion of the Conus venom duct has different peptides (Bingham et al., 1996), venom granules, and pharmacological properties (Endean and Duchemin, 1967) than that extracted from the distal duct. However, the role of this region in venom production remains vague, and no anatomical specialization suggestive of a secretory function has been reported.
TEM images described in this paper reveal that the proximal duct is structurally unique. The high concentration of mitochondria found in the basal region of the epithelial cells must be associated with a high level of energy production, and microvilli greatly increase the apical surface area. Although both features might be associated with secretion, there are no other structural aspects of these cells that support this idea.
In contrast, the above structural features suggest that this tissue mediates active transport. In particular, the epithelium of the proximal venom duct is strikingly similar to the proximal tubule of the mammalian kidney. The latter epithelium absorbs glucose and amino acids along with sodium through apical microvilli and actively transports sodium into the bloodstream across the basal surface with energy provided by numerous mitochondria. Although the nature of the postulated transport process in the proximal venom duct is unknown, it is possible that this epithelium functions in a similar manner to take up ions and small organic molecules (e.g., amino acids, fatty acids) from the duct lumen as part of a recycling mechanism. An attractive source for these molecular substrates is the prominent cellular debris in the lumen.
The distal venom duct
Few venom granules or obvious precursor materials exist in the lumen of the proximal duct, and this suggests that venom granules are manufactured elsewhere, presumably in more anterior duct regions. Numerous venom granules characterize the entire distal venom duct, and they are found both within epithelial cells and in the lumen. Over the length of the distal duct, the relative abundance of granules changes from primarily intracellular proximally to extracellular distally. Cellular debris is also abundant in the lumen of the initial segment of the distal duct (i.e., the more posterior region).
These features strongly suggest that venom granules originate in the distal venom duct. The epithelial cells of this region probably express the genes encoding specific peptides, and granules appear to be assembled in the same cells. The disorganized nature of this epithelium, the poorly defined cell membranes, and the cellular debris in the lumen are all consistent with holocrine secretion, as suggested by previous authors (Halstead, 1988; see also dos Santos et al., 2000). Thus, the abundant cellular debris in the lumen of the proximal duct probably derives from this distal secretory region.
Our analysis of C. californicus contrasts somewhat with work on C. magus by Endean and Duchemin (1967) in which "immature" venom granules were identified within the distal duct and extremely large "mature" granules (up to 20 µm long and 5 µm wide) were found in the proximal duct. On the basis of the presumed developmental progression, those authors suggested that granules did not originate in the posterior duct. Although we agree with that conclusion, we find few granules of any sort in the proximal duct and no extremely large granules anywhere in C. californicus. Reasons for these differences are not clear, and subcellular anatomical studies on other species are inadequate to permit a meaningful comparative analysis.
The connection between the venom duct and proboscis
An important contribution of this paper is the description of the passageway taken by the venom duct into the pharynx (see Fig. 5C). Two features of this region are likely to be important to the manufacture and delivery of venom.
First, the venom duct narrows to channels only 10 µm or less in width as they cross the pharyngeal wall. This appearance can last through 100 µm of serial sections cut parallel to the long axis of the passage. This suggests that the passage is not a cylindrical duct, but rather a flat channel that is much larger in one dimension than the other. The course of these channels becomes somewhat convoluted after the bifurcation, making it impossible to capture the entire length across the pharynx in a single section. Thus, the number of such channels is uncertain, but there are at least two. These channels merge once again upon reaching the lumen of the pharynx, and presumably this is where venom enters the pharynx in vivo (* in Fig. 5A).
The small diameter and considerable length of the fine connecting channels suggest that venom ejection is not likely to involve high flow rates or large volumes. Possibly the cilia lining the distal portions of the fine channels are involved in transport of venom into the pharynx (see also below). A very narrow opening into the pharynx and an associated ciliated tuft were also described in C. magus (Endean and Duchemin, 1967), but the passage within the pharyngeal wall was not elaborated upon.
The second important feature is the unique type of epithelial cell that surrounds the fine duct channels over their course through the pharyngeal wall. These metachromatic cells, which we call purple cells, appear in the wall of the duct just before it enters the pharynx, and they line the passageway to the point at which this epithelium meets the infolded pharyngeal epithelium (see Fig. 5B). The specialized nature and localization of the purple cells suggest that they play a critical role in the production or delivery of venom. It is possible that these cells are secretory and contribute a specific component to the final venom for ejection. Such components could be peptides not found elsewhere in the venom duct (Bingham et al., 1996), enzymes such as phospholipases (McIntosh et al., 1995), or simply mucus.
Alternatively, the purple cells could play an active role in "simplifying" venom from the duct in preparation of the final material to be injected into a victim. Venom "milked" from living Conus of several species contains a much simpler complement of peptidic toxins than that found in crude venom isolated from the duct (Hopkins et al., 1995; Bingham et al., 1996). Our anatomical data suggest that some kind of granule breakdown or sorting process may occur in the fine channels of the venom duct, because at the final branching point, large blue granules are replaced by smaller, clear ones (Figs. 4D–F). Purple cells might selectively take up materials or release specific products that act to break down or modify the granules. In turn, this action could lead to simplification of the peptide complement of the final venom. The purple-cell zone may well represent the site of this process, but the underlying biochemical mechanisms remain to be elucidated.
The radular sac and teeth
As demonstrated in this paper, radular teeth of C. californicus are filled with a rich assortment of granules and other material. This is true for teeth well within the short arm of the radular sac as well as those that are ready to be passed to the proboscis. Examination of non-fixed teeth confirms the presence of granular material. Morphological similarities with material in the lumen of the venom duct make it plausible that peptide toxins exist in both places. This idea is directly supported by mass spectrometry analysis. Thus, C. californicus teeth in the radular sac appear to be "pre-loaded" with peptidic components of venom before being passed to the proboscis for use in hunting. This hypothesis obviously contrasts with the ideas discussed in the Introduction.
A summary model for venom manufacture and delivery in Conus californicus
Profound ultrastructural differences between the proximal and distal portions of the venom duct suggest major functional differences for these regions. We propose that peptide toxins are synthesized and assembled into venom granules within the epithelial cells of the distal venom duct, and that the contents of these cells are discharged into the lumen of the duct, probably by holocrine secretion. Cellular debris resulting from this process would be recycled through an uptake pathway in the proximal venom duct. If so, the muscular bulb may play some role in the reuptake process.
Venom granules are extremely abundant in the lumen of the anteriormost region of the distal venom duct. However, they are prevented from free passage into the pharynx by the network of specialized narrow channels that connects the duct and the pharynx. Slow and limited transport of venom probably occurs across these channels, and some material may be directed into the radular sac, where teeth can be loaded with at least some venom components. This postulated transport need not involve a complex mechanism, and a specialized ciliary tract in a typhlosole-like structure could easily move the requisite material along the highly folded pharyngeal epithelium and into the radular sac.
During the slow passage of venom through the fine duct channels in the purple-cell zone, the granule content and peptide complement of the crude venom could be simplified as hypothesized above. This simplification would appear to be a requirement for in vivo preparation of the final venom for injection. This important feature is difficult to explain if large volumes of material from the anterior venom duct are quickly passed into the pharynx.
Finally, our model is compatible with the possibility that venom injection involves more material than could be contained in the lumen of a single tooth. We are unaware of quantitative data demanding this capability for any Conus species, but it seems likely to be true. Neuronal control over the richly innervated pharyngeal musculature might permit sufficient dilation of the fine duct channels to move a bit of extra venom into the pharynx once the pre-loaded tooth was passed from the radular sac. This bolus of venom could be carried along at the base of the tooth to the tip of the proboscis for final ejection. In this case, if the extra venom were limited to the material of unique granular content contained in the narrow channels, it could still retain a simplified peptide composition.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
|
Literature Cited |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionLiterature Cited |
|---|
This article has been cited by other articles:
|
|
 |
|
  J. Stewart and W. F. Gilly Piscivorous Behavior of a Temperate Cone Snail, Conus californicus Biol. Bull., October 1, 2005; 209(2): 146 - 153. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. A. Jakubowski, W. P. Kelley, J. V. Sweedler, W. F. Gilly, and J. R. Schulz Intraspecific variation of venom injected by fish-hunting Conus snails J. Exp. Biol., August 1, 2005; 208(15): 2873 - 2883. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
