Biol. Bull. 211: 275-285. (December 2006)
© 2006 Marine Biological Laboratory
Peritrophic Membrane of the Penaeid Shrimp Sicyonia ingentis: Structure, Formation, and Permeability
Gary G. Martin*,
Rebecca Simcox,
Aivan Nguyen and
Amaiak Chilingaryan
Department of Biology, Occidental College, Los Angeles, California 90041
* To whom correspondence should be addressed. E-mail: Gmartin{at}oxy.edu
 |
Abstract
|
|---|
Peritrophic membranes (PTMs) are secreted acellular layers that separate ingested materials from the gut epithelium in a variety of invertebrates. In insects and crustaceans, PTMs are produced in the midgut trunk (MGT, or intestine), but the MGT in decapod crustaceans, unlike that of insects, is not involved with digestion or absorption of food. We demonstrate that the PTM in the penaeid shrimp Sicyonia ingentis is similar to that in other crustaceans that have been studied and is primarily composed of chitin. The lectin WGA binds only to the PTM and glycocalyx along the microvilli of the midgut cells, which is consistent with the suggestion that the chitin is synthesized along the microvilli. The PTM is only permeable to inert particles smaller than 20 nm. We also describe the secretion of granules, which fill the apices of the epithelial cells, into the ectoperitrophic space. Although their function is not clear, they do not contribute to the PTM.
Abbreviations: MGT, midgut trunk • NAG, N-acetyl-D-glucosamine • PTM, peritrophic membrane • WGA, wheat germ agglutinin
|
Introduction
|
|---|
Peritrophic membranes (PTMs) are secreted acellular layers that separate ingested materials from the gut epithelium in a variety of invertebrates (Wigglesworth, 1930; Forster, 1953; Georgi, 1969; Dales and Pell, 1970; Peters, 1992; Lehane, 1997), and they persist as wrappers around fecal pellets. They are generally considered to be composed of a meshwork of chitin fibrils embedded in a matrix of proteins, proteoglycans, and mucopolysaccharides (Georgi, 1969; Richards and Richards, 1977; Eisemann and Binnington, 1994; Tellam et al., 1999; Merzendorfer and Zimoch, 2003), although much of this work was performed on insects. It has been suggested that PTMs protect the gut lining from mechanical abrasion, establish functional subdivisions of the gut lumen to facilitate digestion, prevent pathogens and nonspecific materials from binding to epithelial cells, and serve as an ultrafilter to regulate the exchange of nutrients and digestive enzymes (Peters, 1992; Brunet et al., 1994). However, the physiological significance of PTMs deserves further attention (Richards and Richards, 1977).
Most crustaceans, but perhaps not all (see Komura and Yamamoto, 1968; Miyawaki and Taketomi, 1984; To et al., 2004), produce a PTM in the midgut trunk (MGT; also called intestine) and possibly in the digestive gland as well (Forster, 1953; Gauld, 1957; Pillai, 1960; Dall, 1967; Talbot et al., 1972; Quaglia et al., 1976; Lautenschlager et al., 1978; Mykles, 1979; Holliday et al., 1980; Johnson, 1980; Avtsyn and Petrova, 1986; Lovett and Felder, 1990b; Tsuda and Nemoto, 1990; Hansen and Peters, 1997/98; Herrera-Álvarez et al., 2000; De Jong-Moreau et al., 2000; Halcrow, 2001). These PTMs are simple fibrillar layers formed by delamination from the brush border of the epithelial cells (Georgi, 1969). The specific ways that the components of the PTM may be released from these cells have not been clearly established (Brunet et al., 1994). Chitin may be synthesized on the surface of microvilli as described in insects (Ruiz-Herrera and Martinez-Espinoza, 1999; Cohen, 2001; Merzendorfer and Zimoch, 2003), or released from secretory granules from the epithelial cells as described in the calanoid copepod Centropages typicus (Brunet et al., 1994). The purposes of this paper are to (1) compare the structure, basic composition, and permeability of the PTM in Sicyonia ingentis, the ridgeback prawn, with that from other species; (2) describe labeling studies with wheat germ agglutinin (WGA) to support the suggestion that the chitinous part of the PTM is produced along the microvilli; and (3) describe morphological stages in the secretion of granules from the epithelial cells of the MGT.
|
Materials and Methods
|
|---|
Collection and maintenance
Ridgeback prawns, Sicyonia ingentis Burkenroad 1938, were collected by otter trawls in 80 fathoms (160 m) of water off the coast of Palos Verdes and Santa Barbara, California. They were maintained in seawater aquaria at 16 °C and fed pieces of squid. Shrimp made 0–2 fecal pellets each day. To obtain peritrophic membranes (PTMs) free from ingested materials, shrimp were housed individually in 5-1 plastic tubs for 1–3 days without food, and their water was changed daily.
Morphology of midgut trunk and peritrophic membrane
For basic morphological studies, midgut trunks (MGTs) from 30 animals were removed and cut into thirds. Pieces from the anterior, middle, and posterior regions were further cut into four pieces; one piece was processed for histological examination and three pieces were processed for electron microscopy. For histological examination, pieces were fixed for 24 h in Davidson’s fixative (330 ml 95% ethanol, 220 ml 100% formalin, 115 ml glacial acetic acid, and 335 ml distilled water), dehydrated in 100% ethanol, and infiltrated and embedded in JB-4 plastic. Sections (0.5 µm) were stained using routine procedures (Humason, 1972) for hematoxylin and eosin, alcian blue, and bromphenol blue, and a Sigma kit (395B) was used for Periodic acid-Schiff. Sections were examined by brightfield light microscopy (LM).
For electron microscopy, tissues were fixed for 3 h in 2.5% glutaraldehyde in 0.1 mol l–1 sodium cacodylate, pH 8.1, containing 12% glucose. Tissues were rinsed in buffer (0.1 mol l–1 sodium cacodylate), postfixed for 1 h in 1% OsO4 in 0.1 mol l–1 sodium cacodylate, and stained en bloc for 1 h in 3% uranyl acetate in 0.1 mol l–1 sodium acetate buffer. After dehydration in a graded series of ethanol, two samples from each region of each shrimp were processed for transmission electron microscopy (TEM), and the remaining sample was processed for scanning electron microscopy (SEM). Samples for TEM were infiltrated and embedded in Spurr’s (1969) plastic. Thick (0.5-µm) sections were stained with methylene blue and examined by LM. Epithelial cells in these sections displayed morphological changes, which we grouped into three stages. A minimum of five blocks from each section of the MGT showing each stage were thin-sectioned (70 nm), stained with lead citrate, and examined by TEM (Zeiss EM 109). For SEM, dehydrated tissues were either dried by the critical-point method in a Denton DCP-2 apparatus or soaked in hexamethyldisilazane (HMDS Pella 18605) for 10 min and air-dried. Specimens were mounted on stubs, coated with gold, and examined in a Cambridge Stereoscan 360 SEM.
To determine whether there was a difference in the morphology of epithelial cells in the MGT of shrimp given access to food or starved for up to 3 days, pieces of the anterior MGT from five animals starved 3 days were processed and examined by TEM. Sections through these MGTs were compared with cells from animals processed for basic morphology, which had access to food.
En face views of intact PTMs were obtained by collecting clean PTMs from starved animals, slicing them longitudinally to produce a flattened sheet, picking them up on EM grids, and examining them with TEM. Although these images are not as detailed as the shadowed preparations presented by Georgi (1969), they are adequate for visualizing the weave of the presumably chitinous filaments.
Tests for chitin and N-acetyl-D-glucosamine
A basic test for the presence of chitin relies on the fact that it is insoluble in overnight or longer incubations in concentrated solutions of KOH at room or elevated (150 °C) temperatures. This treatment removes the associated proteins, and the residual chitosan dissolves in concentrated HCl or H2SO4 (see Peters, 1992).
To test for the presence of chitin and other materials in the PTM, 5-mm-long pieces of PTMs from five shrimp were soaked in the following solutions: chitinase (2.5 units/ml, Sigma C7809), protease (55 units/ml, Sigma P6911), elastase (100 units/ml, Sigma E0258), lipase (1140 units/ml, Sigma L1754), and collagenase (1000 units/ml, Sigma IV-S 1889). Tissues were observed hourly for the first 6 h and finally after 24 h.
The presence of N-acetyl-D-glucosamine (NAG) has been used as one of the most specific tests for chitin even though this molecule is a component of other molecules (Monsigny et al., 1979; Peters and Latka, 1986). To test for NAG in PTMs, PTMs were fixed briefly (15 min) in Davidson’s fixative. Tissue was washed in Amersham’s phosphate buffered saline (AmPBS: NaCl 8 g, KCl 0.20 g, KH2PO4 0.2 g, Na2HPO4 1.14 g; per liter) and incubated for 1 h in the dark in FITC-linked succinylated WGA (wheat germ agglutinin, i.e., lectin from Triticum vulgarus; Vector FL 1021S; 25 µg/ml AmPBS). After a wash in AmPBS, samples were viewed by fluorescence microscopy.
To localize NAG at the TEM level, tissues were fixed for 20 min in 4% paraformaldehyde and 1% glutaraldehyde in AmPBS, dehydrated, and embedded in LR White (Pella 18181). Thin sections were collected on grids and floated on drops of blocking buffer (AmPBS with 0.5% bovine serum albumin and 50 mol l–1 glycine; Hemphill et al., 2004) for 1 h at room temperature. Sections were then washed in AmPBS, incubated 1 h with gold-labeled WGA (Sigma L1894) diluted 1:50 with AmPBS, rinsed in buffer, and examined by TEM. Two controls were used at both the LM and TEM levels. First, samples were processed through the same solutions as experimental tissues except that they were not incubated with WGA; second, a sample of WGA was mixed with an equal volume of 0.5 mol l–1 NAG or 10 mmol l–1 triacetyl chitotriose before the samples were stained with WGA, as described by Peters and Latka (1986).
Presence of protein in isolated peritrophic membranes
The concentration of protein in PTMs was determined following the procedure of Peters (1992). Clean PTMs from 3-day-starved shrimp were collected and stored at 4 °C in distilled water containing 0.5% sodium azide. Tissue was lyophilized, weighed, and soaked in an aqueous solution of 5% sodium dodecyl sulfate and 8 mol l–1 urea at room temperature for 6 h. Insoluble material was pelleted by centrifugation, and the concentration of proteins in the supernatant was determined using a Bradford assay with albumin (BSA Sigma A-7888) as a standard.
Permeability of peritrophic membranes
The ability of two dyes, methylene blue (MW 373.9 Da) and Evans blue (MW 960.8 Da), to penetrate fully formed PTMs (i.e., those free in the lumen of the MGT and not attached to the epithelium) was observed by using a dissecting microscope. The ends of a PTM were tied closed, and the tube was laid across a depression slide. Dye solutions were added to both sides of the PTM until the tube was just covered. The PTM appeared as a light strip separating the darker dye solution, and the ability of the dyes to move into the endoperitrophic space was followed on 10 preparations examined at 15-min intervals. The same procedure was employed for solutions containing colored particles of known size, including India ink (Higgins; particles 
500 nm) and dyed latex beads (Sigma L3530 500 nm; L1280 100 nm; L1148 55 nm).
For tracers in clear solutions, such as lanthanum nitrate (<2 nm) and cationic ferritin (Sigma F7879; 10–20 nm), TEM was required to determine whether the tracers were able to penetrate the PTM. For these experiments, MGTs were dissected from animals and only those that contained a fully formed PTM were used. A 27-gauge needle attached to a 1-ml syringe filled with test particles was threaded into one end of the PTM, and a thread was used to tie off the PTM and surrounding MGT at that end. The addition of a small amount of methylene blue to the tracer solution proved useful in determining when the injected solution had filled the lumen of the PTMs. Fluid was gently injected into the lumen of the PTM until it was filled, and then the free end of the tube was also tied shut. Five preparations with each tracer were incubated at 15° C for 1 h and then fixed and processed as described for TEM, except that the uranyl acetate step was omitted. Unstained sections were cut from the middle of each preparation and examined by LM and TEM.
Presence of a normal flora
TEM and SEM were used to look for the possible presence of bacteria bound to the brush border of the epithelial cells and in the ectoperitrophic space (between PTM and MGT epithelium). In addition, five shrimp with a fully formed PTM were used for DAPI staining. In each of these animals the fully formed PTM in the lumen of the MGT was removed and the wall of the MGT was cut longitudinally, pinned flat, and fixed in glutaraldehyde solution for 15 min. After a rinse in AmPBS, the MGT was stained with DAPI (1 µg/ml of AmPBS), washed, and examined by fluorescence microscopy.
|
Results
|
|---|
Morphology of the peritrophic membrane
Figure 1, a histological section through the midgut trunk (MGT) of Sicyonia ingentis, shows the single layer of epithelial cells and multiple layers of peritropic membrane (PTM) in the lumen of the gut. Each PTM was eosinophilic and contained carbohydrates (PAS positive) but not acidic mucopolysaccharides (alcian blue negative). PTMs did not stain with bromphenol blue for protein, and a Bradford assay indicated that only about 3.5% of total dry weight was soluble protein. PTMs fluoresced when stained with FITC-conjugated wheat germ agglutinin (WGA), whereas nonstained, control PTMs did not. The addition of N-acetyl-D-glucosamine (NAG) or triacetyl chitotriose to the WGA prior to staining nearly eliminated the fluorescence.
View larger version (166K):
[in this window]
[in a new window]
|
Figures 1–4. (1) Light micrograph (LM) of cross-section through midgut trunk (MGT) showing multiple layers of peritrophic membrane (PTM; arrows) in gut lumen (LU). Note folded epithelium (E), basal lamina (BL), and outer layer of connective tissue and muscle (CT). Scale bar = 100 µm. (2) Scanning electron micrograph (SEM) of inner surface of MGT with PTM present in the upper left portion and removed from the lower right to expose the folded epithelium (E). Scale bar = 500 nm. (3) Transmission electron micrograph (TEM) showing the apex of epithelial cells with a PTM (arrows) along the tips of the microvilli (MV) and another PTM (arrows) recently detached in the gut lumen. Scale bar = 5 µm. (4) TEM of cross-section through a region of fully formed PTM free in the MGT lumen. Note the filamentous nature of the layers of PTM separated by narrow spaces (S). Scale bar = 500 nm.
|
|
Figure 2 shows an SEM view of the luminal surface of the MGT covered in the top half of the picture by a continuous PTM. In the bottom of the figure, the PTM was dissected away to reveal the folds of the epithelium. When examined by TEM (Fig. 3), a thin PTM lies along the tips of the microvilli. PTMs that have detached from the epithelium and lie in the lumen of the MGT are composed of 1–10 thin layers, each the same as those bound to the microvilli. Each layer appears filamentous and merges in places with adjacent layers (Fig. 4). The surface of a fully formed PTM facing the endoperitrophic space (such as in Fig. 2) was typically covered with debris, including bacteria, whereas the outer surface (facing the ectoperitrophic space) was clean. Sections cut tangential to the surface of a PTM often showed a grid pattern that labeled with WGA (Fig. 5). The grid pattern can also be seen in en face views of intact PTMs examined by TEM (Fig. 6). Determined from at least 100 measurements, the diameter of each space in the grid ranged between 170 and 200 nm, the center-to-center distance was between 170 and 250 nm, the diameter of microvilli (Fig. 7). was between 120 and 150 nm, and the center-to-center distance between adjacent microvilli was 190 to 250 nm. The grid pattern was not always observed even in the same preparation, and was not observed in 10 fecal pellets judged to be more than 2 h old.
View larger version (175K):
[in this window]
[in a new window]
|
Figures 5–9. (5) TEM of a section cut tangential to the surface of the PTM showing fibers in a grid pattern labeled with gold-labeled wheat germ agglutinin (WGA; arrow). Scale bar = 500 nm. (6) TEM of intact PTM collected on an EM grid showing its honeycomb structure. Scale bar = 350 nm. (7) TEM showing cross-section through microvilli at apex of epithelial cell allowing for comparison of spacing with pattern of PTM seen in Figure 6. Scale bar = 350 nm. (8) TEM of apex of epithelial cells showing gold-labeled WGA (arrow) bound to the PTM and microvilli but not to the electron-dense secretory granules (G) or smaller vesicles (V). Scale bar = 500 nm. (9) TEM of unstained epithelial cells from the MGT of a shrimp that had lanthanum nitrate injected into its endoperitrophic space. Note the deposits of lanthanum (arrows) in the spaces between cells at a level adjacent to the nucleus (N), which is about 10 µm from the apical surface. Note the cytoplasmic granules (G) and the unique particles (P) attached to the nuclear envelope. Scale bar = 1 µm.
|
|
Does the peritrophic membrane contain chitin and protein?
PTMs removed from the MGT or collected as fecal pellets from shrimp provided with food typically contained ingested materials. Shrimp not given food for 1–3 days continued to produce and release PTM that were relatively clean. No differences were observed in the morphology of the PTM or the epithelial cells from shrimp in these two groups when examined by LM and TEM. PTMs were insoluble in KOH, and they dissolved in concentrated but not dilute HCl and H2SO4. In addition, PTMs dissolved in chitinase, whereas protease, elastase, lipase, and collagenase had no noticeable effect.
A third test for NAG is the overall fluorescence of PTMs incubated with succinylated WGA-FITC, and the labeling of the PTM with this lectin in thin sections examined by TEM (Fig. 8). There was no observed autofluorescence in the unstained samples, and the labeling of the PTM was nearly eliminated when NAG or triacetyl chitotriose was added to the lectin solution before treating the sections. WGA also labeled the brush border of the MGT epithelium, and TEM showed that NAG was localized to the sides and tips of the microvilli. No organelles in the apical cytoplasm were labeled by WGA, and particular attention was given to the small vesicles (50–100 nm) and large electron-dense secretory granules (see Fig. 8).
Permeability of peritrophic membrane
Shrimp PTMs were penetrated by methylene blue and Evans blue almost immediately, but during a 4-h test they were impermeable to India ink (particles 500 nm), and latex beads with diameters ranging down to 55 nm. Examination of TEM sections showed that cationic ferritin (10–20 nm) was not able to penetrate the PTM. LaNO3 (<2 nm), however, passed through the PTM and penetrated between adjacent epithelial cells past the apical region containing the electron-dense secretory granules to the level of the nucleus in all five samples examined (Fig. 9).
Morphological changes to the epithelial cells of the midgut trunk and the production of the peritrophic membrane
The basic morphology of epithelial cells lining the MGT in S. ingentis has been described (Martin and Chiu, 2003). The present study found that the epithelial cells from the anterior, middle, and posterior regions of the MGT all showed the same morphology at any given time. The cytoplasm in the apical third of these cells underwent a series of morphological changes involved with the formation, hydration, and secretion of cytoplasmic granules. In stage 1 (Fig. 10), the apical cytoplasm was filled with granules that were eosinophilic and did not stain with PAS, bromphenol blue, or WGA. In sections, the granules (0.4–0.6 µm) had a circular to ovoid outline, and the content was homogeneous and electron-dense. Between the granules were a small number of mitochondria, Golgi bodies, and vesicles (50–100 nm). A membrane-bound space containing membranous whorls was typically seen apical to the nucleus. A continuous brush border lined the top of the epithelium. When the luminal surface of the MGT was examined by SEM, individual microvilli were rarely observed. Instead, clusters of adjacent microvilli were covered by a material (Fig. 11) that increased in amount until it formed a continuous layer obscuring the brush border (Fig. 12). TEM of the microvilli revealed a glycocalyx that adhered to their sides and tips and appeared identical to the developing PTM seen along the tips of the microvilli (Fig. 13).
View larger version (202K):
[in this window]
[in a new window]
|
Figures 10–13. (10) TEM of apices of epithelial cells from shrimp in stage 1 showing electron-dense granules and multivesicular bodies (MVB) above the nuclear (N) region of each cell. A PTM is forming at the tips of the microvilli (MV). Scale bar = 2 µm. (11) SEM of MGT epithelium showing microvilli clustered (C) together by material considered to be an initial stage in formation of PTM. Scale bar = 5 µm. (12) SEM of folded epithelium surface from stage 1 shrimp. Cells on the left are covered by an incomplete PTM similar to that seen in Figure 12. The PTM covering cells on the right is considered more developed because it is smooth and continuous. Scale bar = 50 µm. (13) TEM showing glycocalyx material (arrows) on sides and tips of microvilli (MV); this material is identical to the PTM lying along the surface of the microvilli. Scale bar = 100 nm.
|
|
In stage 2, a single, thin (100 nm) PTM lay along the tips of the microvilli (Figs. 3 and 14), and the glycocalyx along the sides of the microvilli was less dense. In the apical cytoplasm, the electron-dense granules were less common than in stage 1, presumably because most had hydrated to form electron-lucent vesicles filling the cell apex. Intermediate stages suggest that the granule contents first dispersed to reveal small electron-dense deposits, and with further hydration the entire granule contents became electron-lucent. As the appearance of the granule content changed, adjacent granules fused to produce large, irregularly shaped spaces filling most of each cell apical to the nuclear region. The small vesicles were as common as in stage 1, but large supranuclear multivesicular bodies were rare. SEM views of the MGT showed a nearly continuous PTM (Fig. 15). Occasional small ovoid holes and slits in the PTM may represent areas not completely covered by secreted matrix, or perhaps drying artifacts. Views through these gaps showed that the PTM lay along the tips of the microvilli.
View larger version (194K):
[in this window]
[in a new window]
|
Figures 14–17. (14) TEM of apices of epithelial cells from stage 2 shrimp showing a reduced number of electron-dense granules (G); numerous hydrated, electron-lucent vesicles (V); and presumably intermediate stages with small electron-dense deposits (arrows) inside electron-lucent vesicles. Note the continuous row of microvilli (MV) and developing PTM. Scale bar = 5 µm. (15) SEM of PTM covering the tips of microvilli (MV) in stage 2 MGT. It is continuous except for a few small gaps (arrows), which may be artifacts of drying or incomplete stages in PTM formation. Scale bar = 5 µm. (16) TEM of epithelial cells of a stage 3 shrimp showing large apical expansions (E) from the cells extending past the microvilli (MV). Note the lack of granules and the numerous electron-lucent vesicles (V) filling the cell apices. Scale bar = 5 µm. (17) SEM of cell apices from stage 3 shrimp showing small (S) and large (L) extensions from the cell apices. Other cells show pits (P) where the secretions have lysed, leaving a ring of microvilli surrounding a depressed apical cell surface. Scale bar = 10 µm.
|
|
During stage 3, large electron-lucent extensions of the apical cytoplasm spread beyond the tips of the microvilli (Fig. 16). Only a few electron-dense granules remained in the apical cytoplasm, which was filled with electron-lucent vesicles merging into the large expansion of the cell apex. The brush border, which had been continuous in previous stages, was interrupted by the expansion of the cell apices. When examined by TEM, the microvilli were covered with the same glycocalyx as in stage 1. The material derived from the granules was released into the ectoperitrophic space because the PTM had lifted off the surface of the epithelium and could be readily removed from the MGT without adhesions to the epithelium. Because the PTM did not form a continuous layer during this stage, the apex of individual epithelial cells could be observed by SEM. About 10% of the cells we examined had a complete layer of microvilli across their surfaces. In most of the cells, the microvilli were located around the periphery of either a large (up to 10 µm diameter), spheroid extension of the cell apex, or a pit if the apical secretion had lysed (Fig. 17). To address the possibility that these cell extensions seen both in sectioned samples and by SEM could be artifacts of fixation, additional MGTs were fixed by (1) injecting fixative into the lumen, followed by immersion, (2) slicing some into 1-mm sections to facilitate the rapid penetration of solution, and (3) increasing the osmolarity of the fixative. Regardless of the procedure used, expansions of the apical cytoplasm were observed in shrimp showing the other features of stage 3, but in no other stages.
To determine whether there is a cycle to the formation of the PTM, individual shrimp were placed in separate tubs and inspected every 30 min. The MGTs from 20 shrimp that had recently passed a fecal pellet and from 40 additional animals selected at random times were removed, fixed, and processed for examination by LM and EM. Epithelial cells in the MGT of animals that had just released fecal pellets were routinely in stage 1, as were some (7/40) of the randomly selected animals. Ingested materials in the MGTs of these animals were surrounded by 1–3 layers of PTM. Most of the randomly selected animals showed MGTs, which we classified as stages 2 (21/40) or 3 (12/40). We never observed more than a single PTM along the tips of the apical microvilli, and between 1 and 10 layers of PTM encircled material in the lumen of the MGT. We were not able to determine the length of time required for the epithelium to pass from one stage to the next because only stage 1 can be estimated without dissecting the animals, and a variable number of PTMs may surround subsequent fecal pellets before they are defecated.
Presence of a normal flora
Bacteria were not observed in the ectoperitrophic space or attached to the brush border of the MGT in sections examined by LM and TEM. SEM did not show any bacteria on the external surface of PTMs dissected from MGTs, although bacteria were seen on fecal pellets that had been exposed to seawater for as little as 15 min. Shrimp with PTMs clearly detached from their MGTs were selected for staining with DAPI because they provided the cleanest views of the microvilli of the epithelial surface. No fluorescent bodies of bacterial size were observed along the brush border.
|
Discussion
|
|---|
The peritrophic membrane (PTM) in the penaeid shrimp Sicyonia ingentis is similar to PTMs described in other arthropods (Georgi, 1969; Richards and Richards, 1977; Peters, 1992). It is an acellular sac, composed primarily of chitin. In addition, it is labeled by succinylated wheat germ agglutinin (WGA), a lectin that is considered specific for N-acetyl-D-glucosamine (NAG) and has been used to help identify chitinous structures, especially when employed in conjunction with competitive inhibitors (Monsigny et al., 1979; Peters and Latka, 1986). Chitin is not found alone in nature but is complexed to protein (Rudall and Kenchington, 1973). Only about 3.4% of the total dry weight of the PTM of S. ingentis is SDS-urea soluble protein.
It is generally agreed that PTMs in decapod crustaceans are produced in the midgut trunk (MGT), and layers of fibrous material form between the microvilli (Talbot et al., 1972; Johnson, 1980; Brunet et al., 1994) and delaminate from their tips (Georgi, 1969). Georgi (1969) used TEM to examine shadowed, intact specimens to demonstrate PTMs with fibrillar weaves with random, orthogonal, and hexagonal patterns. In S. ingentis, unshadowed preparations of the PTM show an overlap between the dimensions of the weave pattern and the spacing of the microvilli, as seen in other crustaceans (Peters, 1992; Halcrow, 2001). The microvilli lining the epithelial cells of the MGT in S. ingentis are covered by a glycocalyx, which labels with WGA and appears identical to the PTM seen along their tips. Labeling of the brush border in S. ingentis with WGA agrees with work on insects (Peters and Latka, 1986; Rudin and Hecker, 1989) that suggests that synthesis of the chitinous part of the PTM occurs along the microvilli. It is not clear whether chitin synthase is inserted into plasma membranes and catalyses the polymerization of chitin fibrils in the extracellular space, or if chitin synthase initiates the production of chitin in vesicles, before their fusion with the plasma membrane (Ruiz-Herrera and Martinez-Espinoza, 1999; Cohen, 2001; Merzendorfer and Zimoch, 2003).
All PTMs separate large particulate material in the gut lumen from the gut epithelium. In S. ingentis, bacteria and other particles of similar size were routinely observed in the endoperitrophic space, but none were seen along the brush border when examined using SEM, TEM, and DAPI staining. The absence of a normal flora in S. ingentis agrees with a report by Boyle and Mitchel (1978), who found no bacteria associated with the digestive tract of three species of wood-boring isopods and a wood-inhabiting amphipod. Herrera-Álvarez et al. (2000) showed bacteria in close contact with epithelial cells in a mystacocarid, Derocheilocaris remanei, but similar images in the Decapoda are uncommon. There have been numerous descriptions of normal flora in crustaceans (Lee and Pfeifer, 1977; Alvarez, 1983; Dempsey et al., 1989; Jayabalan and Pillai, 1994; Oxley et al., 2002), but most of these studies did not discriminate between the presence of bacteria in the endoperitrophic space and in the ectoperitrophic space. Within its lumen, the PTM may effectively sequester bacteria and particles of similar size acquired by feeding, but this does not explain the absence of these particles in S. ingentis, where water enters the ectoperitrophic space by anal pumping prior to defecation (Fox, 1952; Pillai, 1960; Dall, 1967; Lovett and Felder, 1990b).
Our observations on S. ingentis suggest that only inert particles less than 20 nm can pass through the PTM, and therefore it is more restrictive than the PTMs of four small crustaceans (Daphnia magnus, Artemia salina, Cyclops strenuous, and Cyclops strenuous) whose PTMs are permeable in the range of 70–327 nm (Hansen and Peters, 1997/98). The values for these crustacean PTMs are larger than effective pore sizes reported in insects, which can be in the range of 1.13 to 5 nm (Peters and Wiese, 1986; Peters, 1992; Eisemann and Binnington, 1994). Are PTMs effective in preventing or retarding the penetration of microbes? It seems likely, as even the largest pore sizes reported for the crustaceans are small enough to restrict the passage of particles in the size range of many microbes. Avtsyn and Petrova (1986) suggested that the PTM lining the MGT of Daphnia magna provides resistance to pathogenic bacteria. Nayduch et al. (2005) reported that houseflies trap the enteropathogen Aeromonas caviae between successive layers of their PTM, and the PTM in larval Trichoplusia ni has been shown to limit infection by baculovirus (Wang and Grandados, 1998). Peters (1992) also presented examples in which PTMs are ineffective in protecting the body, as with bacteria that release enzymes to degrade these wrappers (see also Fuhrman et al., 1992; Shahabuddin et al., 1993). Furthermore, pathogens need not penetrate the PTM to cause damage. Toxins from the bacterium Vibrio parahaemolyticus may pass through the PTM of S. ingentis, causing sloughing of the MGT epithelium (Martin et al., 2004). Formation of a barrier to pathogen penetration in decapod crustaceans is probably only one role of the PTM, and alternative functions deserve further study.
This paper also describes the secretory cycle of the major cell type in the MGT of S. ingentis. Although we originally considered the granules in these cells as a possible source of the chitin in the PTM, we concluded that the granules do not contribute to the chitinous PTM. They do not label with WGA and they are secreted after the PTM has separated from the brush border. In addition, abundant granules, like those in epithelial cells of S. ingentis, are not seen in the MGT epithelium of other decapods including Penaeus astecus (Talbot et al., 1972), Penaeus setiferus (Lovett and Felder, 1990a), Procambarus clarkii (Komuro and Yamamoto, 1968), Macrobrachium rosenbergii (Cooke and Ahearn, 1976), Portunus sanguinolentus (Babu et al., 1989), and Paleomonetes pugio, Panulirus interruptus, Loxorhynchus grandis, and Homarus americanus (pers. obs.).
We propose the following sequence of changes in the secretory cells of the MGT epithelium in S. ingentis. At any given time, cells along the length of the MGT are all in the same stage. During stage 1, the PTM is forming; it does not form a continuous layer, and at the TEM level it appears as a glycocaylx lining the sides and tips of the microvilli. The apical cytoplasm is full of granules that are electron-dense and do not stain with WGA. During stage 2, a thin PTM is formed and lies as a continuous layer along the tips of the microvilli. Cytoplasmic granules hydrate and become electron-lucent. In Stage 3, the PTM lifts off the microvilli. The hydrated granules fuse, expand the cell apices beyond the tips of the microvilli, and release their content into the ectoperitrophic space. The possibility that these apical extensions are artifacts (see Khan and Ford, 1962; de Priester, 1971; Ryerse et al., 1992) was addressed by employing several fixatives on MGTs of different stages. Although we cannot rule out the possibility that these expansions are an artifact of fixation, all cells at all stages showed the same ultrastructure of the mitochondria, nuclear envelopes, and cytoplasm. Only the granules showed the cyclic changes. Further studies are needed to integrate the role of the secretory granules into the functioning of the MGT.
|
Acknowledgments
|
|---|
We thank Captain Mike McCorkle and the crew of R/V Vantuna for collecting the shrimp, Alicia Thompson for use of the SEM at USC, Dr. Renee Baran for use of a fluorescence microscope, Lorie Dwyer for translations, Mabby Nelson for help with measurements, and the Undergraduate Research Center at Occidental College for support to AC, AN, and RS.
|
Footnotes
|
|---|
Received 22 March 2006; accepted 16 October 2006.
|
Literature Cited
|
|---|
Alvarez, R. J. 1983. Frequency and distribution of bacterial flora of Penaeus shrimp. Caribbean J. Sci. 19:43–48.
Avtsyn, A. P., and T. P. Petrova. 1986. Resistance of the alimentary canal of Daphnia magna Straus to the effect of enteropathogenic NAG-Vibrios. Biul. Eksp. Biol. Med. 102:342–345.
Babu, B. R., K. Shyamasundari, and K. H. Rao. 1989. Observations on the morphology and histochemistry of the midgut and hindgut of Portunus sanguinolentus (Herbst) (Crustaceans: Brachyura). Folia Morphol. 37: 373–381.
Boyle, P. J., and R. Mitchel. 1978. Absence of microorganisms in crustacean digestive tracts. Science 200: 1157–1159.[Abstract/Free Full Text]
Brunet, M., J. Arnaus, and J. Mazza. 1994. Gut structure and digestive cellular processes in marine crustacean. Oceanogr. Mar. Biol. Ann. Rev. 32: 335–367.
Cohen, E. 2001. Chitin synthesis and inhibition: a revisit. Pest Manag. Sci. 57: 946–950.[ISI][Medline]
Cooke, W. J., and G. A. Ahearn. 1976. Comparative cell morphology of freshwater and marine shrimp midgut epithelia. Am. Zool. 16: 225.
Dales, R. P., and J. S. Pell. 1970. The nature of the peritrophic membrane in the gut of the terebellid polychaete Neoamphitrite figulus. Comp. Biochem. Physiol. 34:819–826.
Dall, W. 1967. The functional anatomy of the digestive tract of a shrimp Metapenaeus bennettae Racek & Dall (Crustacea: Decapoda: Penaeidae). Aust. J. Zool. 15: 699–714.
De Jong-Moreau, L., M. Brunet, J.-P. Casanova, and J. Mazza. 2000. Comparative structure and ultrastructure of the midgut and hepatopancreas of five species of Mysidacea (Crustacea): functional implications. Can. J. Zool. 78: 822–834.
de Priester, W. 1971. Ultrastructure of the midgut epithelial cells in the fly Calliphora erythrocephala. J. Ultrastruct. Res. 36:783–805.
Dempsey, A. C., C. L. Kitting, and R. A. Rosson. 1989. Bacterial variability among individual penaeid shrimp digestive tracts. Crustaceana 56:267–278.[Medline]
Eisemann, C. H., and K. C. Binnington. 1994. The peritrophic membrane: its formation, structure, chemical composition and permeability in relation to vaccination against ectoparasitic arthropods. Int. J. Parasitol. 24:15–26.[ISI][Medline]
Forster, G. R. 1953. Peritrophic membranes in the Caridea (Crustacea Decapoda). J. Mar. Biol. Assoc. UK 32:315–318.
Fox, H. M. 1952. Anal and oral intake of water by crustacea. J. Exp. Biol. 29:583–599.[Abstract]
Fuhrman, J. A., W. S. Lane, R. F. Smith, W. F. Piessens, and F. B. Perler. 1992. Transmission-blocking antibodies recognize microfilarial chitinase in brugian lymphatic filariasis. Proc. Natl. Acad. Sci. USA 89:1548–1552.[Abstract/Free Full Text]
Gauld, D.T. 1957. A peritrophic membrane in calanoid copepods. Nature 179:325–326.
Georgi, R. 1969. Bildung peritrophischer Membranen von Decapoden. Z. Zellforsch. Mikrosk. Anat. 99:570–607.[ISI][Medline]
Halcrow, K. 2001. Ultrastructural features of the funnel of Gammarus oceanicus (Amphipoda). J. Crustac. Biol. 21:631–639.
Hansen, U., and W. Peters, 1997/98. Structure and permeability of the peritrophic membranes of some small crustaceans. Zool. Anz. 236:103–108.
Hemphill, A., N. Vonlaufen, A. Naguleswaran, N. Keller, M. Riesen, N. Guetg, S. Srinivasan, and F. Alaeddine. 2004. Tissue culture and explant approaches to studying and visualizing Neospora caninum and its interactions with the host cell. Microsc. Microanal. 10:602–620.[ISI][Medline]
Herrera-Álvarez, L., I. Fernández, J. Benito, and F. Pardos. 2000. Ultrastructure of the midgut and hindgut of Derochelocaris remànei (Crustacea, Mystacocarida). J. Morphol. 244:177–189.[ISI][Medline]
Holliday, C.W., D.L. Mykles, R.C. Terwilliger, and L.J. Dangott. 1980. Fluid secretion by the midgut caeca of the crab, Cancer magister. Comp. Biochem Physiol. 67A:259–263.
Humasson, G. 1972. Animal Tissue Techniques. W.H. Freeman, San Francisco.
Jayabalan, N., and D. Pillai. 1994. Penaeid prawns and associated luminous bacteria. Acta Microbiol. Immunol. Hung. 41:415–421.[Medline]
Johnson, P. T. 1980. Histology of the Blue Crab, Callinectes sapidus. A Model for the Decapoda. Praeger Scientific, New York.
Khan, M. R., and J. B. Ford. 1962. Studies on digestive enzyme production and its relationship to the cytology of the midgut epithelium in Dysdercus fasciatus Sign. (Hemiptera, Pyrrhocoridae). J. Insect Physiol. 8:597–608.
Komuro, T., and T. Yamamoto. 1968. Fine structure of the epithelium of the gut in the crayfish (Procambarus clarkii) with special reference to the cytoplasmic microtubules. Arch. Histol. Jpn. 30:17–32.[Medline]
Lautenschlager, K. P., N. K. Kaushik, and J. B. Robinson. 1978. The peritrophic membrane and faecal pellets of Gammarus lacustris limnaeus Smith. Freshw. Biol. 8:207–211.
Lee, J. S., and D. K. Pfeifer. 1977. Microbiological characteristics of pacific shrimp (Pandalus jordani). Appl. Environ. Microbiol. 33:853–859.[Abstract/Free Full Text]
Lehane, M.J. 1997. Peritrophic matrix structure and function. Annu. Rev. Entomol. 42:525–550.[ISI][Medline]
Lovett, D. L., and D. L. Felder. 1990a. Ontogenetic changes in enzyme distribution and midgut function in developmental stages of Penaeus setiferus (Crustacea, Decapoda, Penaeidae). Biol. Bull. 178:160–174.[Abstract]
Lovett, D. L., and D. L. Felder. 1990b. Ontogeny of kinematics in the gut of the white shrimp Penaeus setiferus (Decapoda: Penaeidae). J. Crustac. Biol. 10:53–68.
Martin, G. G., and A. Chiu. 2003. Morphology of the midgut trunk in the penaeid shrimp, Sicyonia ingentis: highlighting novel nuclear pore particles and fixed hemocytes. J. Morphol. 258:239–248.[ISI][Medline]
Martin, G. G., N. Rubin, and E. Swanson. 2004. Vibrio parahaemolyticus and V. harveyi cause detachment of the epithelium from the midgut trunk of the penaeid shrimp Sicyonia ingentis. Dis. Aquat. Org. 60:21–29.[ISI][Medline]
Merzendorfer, H., and L. Zimoch. 2003. Chitin metabolism in insects: structure, function and regulation of chitin synthases and chitinases. J. Exp. Biol. 206:4393–4412.[Abstract/Free Full Text]
Miyawaki, M., and Y. Taketomi. 1984. The microvilli of the midgut epithelium in the freshwater shrimp, Caridina denticulate. Cell Biol. Int. Rep. 8:265–268.[ISI][Medline]
Monsigny, M., C. Sene, A. Obrenovitch, A. C. Roche, F. Delmotte, and E. Boschetti. 1979. Properties of succinylated wheat-germ agglutinin. Eur. J. Biochem. 98:39–45.[ISI][Medline]
Mykles, D. L. 1979. Ultrastructure of the alimentary epithelium of lobsters, Homarus americanus and H. gammarus, and the crab Cancer magister. Zoomorphologie 92:201–215.
Nayduch, D., G. P. Noblet, and F. J. Stutzenberger. 2005. Fate of bacteria, Aeromonas caviae, in the midgut of the housefly, Musca domestica. Invertebr. Biol. 124:74–78.
Oxley, A. P. A., W. Shipton, L. Owens, and D. McKay. 2002. Bacterial flora from the gut of the wild and cultured banana prawn, Penaeus merguiensis. J. Appl. Microbiol. 93:214–223.[Medline]
Peters, W. 1992. Peritrophic Membranes. Springer-Verlag, Berlin.
Peters, W., and I. Latka. 1986. Electron microscopic localization of chitin using colloidal gold labeled with wheat germ agglutinin. Histochemistry 84:155–160.[ISI][Medline]
Peters, W., and B. Wiese. 1986. Permeability of the peritrophic membranes of some diptera to labeled dextrans. J. Insect Physiol. 32:43–49.
Pillai, R. S. 1960. Studies on the shrimp Caridina laevis (Heller). I. The digestive system. J. Mar. Biol. Assoc. India 2:57–74
Quaglia, A., B. Sabelli, and L. Villani. 1976. Studies on the intestine of Daphnidae (Crustacea, Cladocera). Ultrastructure of the midgut of Daphnia magna and Daphnia obtusa. J. Morphol. 150:711–726.
Richards, A. G., and P. A. Richards. 1977. The peritrophic membranes of insects. Annu. Rev. Entomol. 22:219–240.[ISI][Medline]
Rudall, K. M., and W. Kenchington. 1973. The chitin system. Biol. Rev. 48:597–636.
Rudin, W., and H. Hecker. 1989. Lectin-binding sites in the midgut of the mosquitoes Anopheles stephensi Liston and Aedes aegypti L. (Diptera: Culicidae). Parasitol. Res. 75:268–279.[ISI][Medline]
Ruiz-Herrera, J., and C. Martinez-Espinoza. 1999. Chitin biosynthesis and structural organization in vivo. EXS 87:39–53.[Medline]
Ryerse, J.S., J.P. Purcell, R.D. Summers, and P.B. Lavrik. 1992. Peritrophic membrane structure and formation in the larva of a moth, Heliothis. Tissue Cell 24:751–771.
Shahabuddin, M., T. Toyoshima, M. Aikawa, and D. Kaslow. 1993. Transmission-blocking activity of a chitinase inhibitor and activation of malarial parasite chitinase by mosquito protease. Proc. Natl. Acad. Sci. USA 90:4266–4270.[Abstract/Free Full Text]
Spurr, A. 1969. A low viscosity epoxy embedding medium for electron microscopy. J. Ultrastruct. Res. 26:31–43.[ISI][Medline]
Talbot, P., W. H. Clark Jr., and A. L. Lawrence. 1972. Fine structure of the midgut epithelium in the developing brown shrimp, Penaeus aztecus. J. Morphol. 138:467–486.
Tellam, R.L., G. Wijffels, and P. Willadsen. 1999. Peritrophic matrix proteins. Insect Biochem. Mol. Biol. 29:87–101.[ISI][Medline]
To, T.H., T.L. Brenner, M.J. Cavey, and J.L. Wilkens. 2004. Histological organization of the intestine in the crayfish Procambarus clarkii. Acta Zool. 85:119–130.
Tsuda, A., and T. Nemoto. 1990. The effect of food concentration on the faecal pellet size of the marine copepod Pseudocalanus newmani Frost. Bull. Plankton Soc. Jpn. 37:83–90.[Medline]
Wang, P., and R. R. Grandados. 1998. Observations on the presence of the peritrophic membrane in larval Trichoplusia ni and its role in limiting Baculovirus infection. J. Invertebr. Pathol. 72:57–62.[ISI][Medline]
Wigglesworth, V. B. 1930. The formation of the peritrophic membrane in insects, with special reference to the larvae of mosquitoes. Q. J. Microsc. Sci. 73:593–616.
TOP
Abstract
Introduction
