Stem Cells 2003;21:322-336
www.StemCells.com
© 2003 AlphaMed Press
Defining Epithelial Cell Progenitors in the Human Oxyntic Mucosa
Sherif M. Karama,b,
Timothy Straitonc,
,
Wail M. Hassanb,
Charles Philippe Leblondc
a Department of Anatomy, Faculty of Medicine and Health Sciences, UAE University, Al-Ain, United Arab Emirates;
b Department of Anatomy, Faculty of Medicine, Kuwait University, Kuwait;
c Department of Anatomy and Cell Biology, McGill University, Montreal, Canada;
Deceased
Key Words. Gastric epithelium • Gastric gland • Stem cell • Cell differentiation • Cell lineage
Sherif M. Karam, M.D., Ph.D., Department of Anatomy, Faculty of Medicine & Health Sciences, United Arab Emirates University, Al-Ain, PO Box 17666, United Arab Emirates. Telephone: 971-03-703-9493; Fax: 971-03-767-2033; e-mail: skaram{at}uaeu.ac.ae
 |
ABSTRACT
|
|---|
In the human stomach, the oxyntic epithelium includes numerous tubular invaginations consisting of short pits opening into long glands. The pit is lined by pit cells, whereas the gland is composed of three regions: the base, containing zymogenic cells; the neck, containing neck cells; and the isthmus, composed of little known immature cells and of parietal cells, which are also scattered through the neck and base. The aim of this study was to examine the ultrastructure of the immature cells and to determine their relation to mature cells. To do so, normal oxyntic mucosal biopsies from subjects ranging from 20–43 years old were fixed in aldehydes and postfixed in reduced osmium for electron microscopy and morphometric analysis. The immature cells were sorted out into four classes, whose roles were clarified by comparison with the thoroughly investigated mouse oxyntic epithelium. The first class was composed of the least differentiated immature cells, which were rare and characterized by minute, dense, or cored secretory granules and were accordingly named mini-granule cells. Their function was not clarified. The second class consisted of pre-pit cells, which were characterized by few dense mucous granules and give rise to pit cells that ascend the pit wall and, after reaching the luminal surface, die or are extruded. Both pre-pit and pit cells underwent continuous renewal and, therefore, together constituted a renewal system referred to as pit cell lineage. The third class, or pre-neck cells, characterized by cored secretory granules, give rise to neck cells that descend toward the base region and differentiate further into pre-zymogenic cells, which finally become zymogenic cells. The latter eventually degenerate and die. Thus pre-neck cells and their progeny constitute a renewing system, designated zymogenic cell lineage. The fourth class, or pre-parietal cells, characterized by long microvilli and few tubulovesicles, differentiate into parietal cells that descend along the neck and base regions and eventually degenerate and die. Pre-parietal and parietal cells represent a renewing system referred to as parietal cell lineage. While the origin of the last three classes of progenitor cells has not been elucidated, it is likely that they arise either from an unidentified multipotential stem cell, possibly the mini-granule cell itself, or from the mitotic activity of pre-pit and pre-neck cells. In conclusion, the human oxyntic epithelium is composed of continually renewing cells organized in distinct cell lineages.
|
INTRODUCTION
|
|---|
In the human stomach, the luminal surface of the oxyntic mucosa displays numerous orifices that lead to tubular invaginations called foveolae or pits, at the bottom of which open one or more glands. Each gland is divided into three regions: isthmus, neck, and base (Fig. 1
). Four types of epithelial cells defined by features summarized in Table 1 are distributed along the pit and gland regions [1–4]. Two of these cells are mucus secreting, namely, the surface mucous cell or pit cell that lines the pit, and the mucous neck cell that lines the neck. The third cell type is the pepsinogen-secreting zymogenic cell that lines the base, while the fourth is the acid-producing parietal cell, which is scattered through the gland (Fig. 1). In addition, the epithelium includes poorly characterized cells referred to as undifferentiated or immature, which are thought to give rise to mature parietal and mucous cells [1]. The goal of this study was twofold: first, to reexamine the immature cells of the human oxyntic epithelium by the electron microscope, and second, to examine their role in the origin of mature cells.
View larger version (37K):
[in this window]
[in a new window]
|
Figure 1. Human oxyntic mucosa drawn in the light microscope. The pit is depicted in continuity with a single gland composed of isthmus, neck, and base (but the pit could be in continuity with two or more glands). The pit is lined by pit cells. The apical region of these cells includes dense mucous granules, which are loosely scattered in low pit but packed close to the apical membrane in mid and high pit, while gradually increasing in number up to the surface. There, the number of granules per cell decreases sharply. The dark portion of a cell at top left is due to its pyknotic nucleus, while the dark granules in a cell at top right are indicative of their undergoing disintegration. The gland isthmus is mainly composed of immature cells, four classes of which are distinguished, namely, mini-granule cells (class 1), pre-pit cells (class 2) predominating in the high isthmus, pre-neck cells (class 3) predominating in the low isthmus, and pre-parietal cells (class 4) found throughout the isthmus. The gland neck is characterized by neck cells displaying cored mucous granules. The gland base is composed of zymogenic cells. Parietal cells are dispersed throughout the gland but are lacking in the pit. There is a scattering of enteroendocrine cells. Finally, the following cells can be in close contact: pre-pit and pit cells at the pit-isthmus boundary; pre-neck and neck cells at the isthmus-neck boundary; and neck, pre-zymogenic, and zymogenic cells at the neck-base boundary. [Drawn by Yves Clermont and reproduced with his permission].
|
|
In rodents, much is known about the origin of mature cells [5–7]. The sequence of events is particularly clear in the mouse. First, one of the immature cells, referred to as the granule-free cell and located in the gland isthmus, functions as the multipotential stem cell of the epithelium in giving rise to three progenitor cell types designated "pre-pit," "pre-neck," and "pre-parietal" cells [8]. Radioautography at successive times after 3H-thymidine injection revealed that, first, pre-pit cells yield pit cells that migrate upward along the pit wall to the free gastric surface [9]. Second, pre-neck cells give rise to neck cells that migrate through the gland neck down to the neck-base border where they transform into the so-called prezymogenic cells, which then become zymogenic cells [10], and third, pre-parietal cells yield parietal cells, which migrate from the gland isthmus to either the gland base or pit [11].
In contrast, little is known about the origin of mature cells in humans. However, radioautography of cultured gastric biopsies exposed to 3H-thymidine [12, 13] indicated that, about 3 days after 3H-thymidine addition to the culture [14, 15], pit cells migrated from the pit bottom up the pit walls to the luminal surface [16], where 2.4% of them died [17], while the others were shed to the gastric lumen at the rate of 5.5 x 105 cells per minute [18]. Of the other mature cells, neck cells have been found to divide, so some investigators propose that they are self-renewing and even function as the stem cell of the gastric epithelium [3, 19, 20], while others believe that neck cells come from immature cells [1]. There is also controversy on the origin of parietal cells, as reviewed elsewhere [21]. Finally, with regard to the origin of zymogenic cells, some propose that they renew themselves by mitosis [19, 20], while others, having noticed their presence at the neck-base border of cells containing secretory granules intermediate to those of neck and zymogenic cells, conclude that zymogenic cells arise from the transformation of neck cells [22, 23], a theory rejected by other investigators [24, 25]. The cells referred to as endocrine or enteroendocrine were considered in this investigation.
Ideally, the role of human immature cells in the production of mature cells could be unequivocally clarified by the use of timed 3H-thymidine radioautography, as done in the mouse. However, the efficiency of 3H-thymidine incorporation into cultured biopsy is not good enough to trace the development of mature cells over long periods, nor would it be ethical to inject humans with radioactive thymidine to label progenitor cells and follow their differentiation in vivo. Therefore, we decided to carry out a systematic comparison between the poorly known human epithelium and the thoroughly understood mouse epithelium. In practice, the approach consists of two steps as follows.
The first step, as presented in the Results section, consists of the electron microscopic analysis of the oxyntic epithelium with particular attention to immature cells, associations of mature and immature cells, and degenerating cells. The second step, as presented in the Discussion section, consists of systematically comparing the observed features of the human oxyntic epithelium with the relevant mouse features, whose behavior has been previously demonstrated with the help of timed 3H-thymidine radioautography. The comparison has made it possible to obtain reasonably trustworthy evidence on the dynamic features of cells in the human epithelium.
|
MATERIALS AND METHODS
|
|---|
Human Tissues
Normal gastric mucosal biopsies were obtained from the body region of the stomach of informed human subjects ranging from 20–43 years of age. In Kuwait, biopsies were obtained by Dr. Dia Mamoon, and in Montreal, by Dr. Gary Wild after approval by the Ethics Committee of the Montreal General Hospital.
Ultrastructural Studies
Immediately after removal of the stomach samples, they were fixed in 0.1 M sodium cacodylate buffer containing 2.5% glutaraldehyde and 2% paraformaldehyde. After a 3-hour fixation, the tissues were washed several times in 0.1 M cacodylate buffer, postfixed in 1% osmium tetroxide partially reduced by 3% potassium ferrocyanide, dehydrated in graded alcohols, and then infiltrated and embedded in Araldite (TAAB Lab Equipment Ltd.; Birkshire, UK). Semithin (0.5 µm thick) sections were cut and stained with toluidine blue for light microscopy. If these sections showed that pits and glands were not oriented in a direction perpendicular to the luminal surface of the stomach, the tissue block was trimmed for reorientation in this direction. Ultrathin sections were cut, mounted on copper grids, and stained with uranyl acetate and lead citrate. The sections were first examined in a Phillips 400 electron microscope (Phillips Ltd.; Eindhoven, The Netherlands) in Montreal and later in a Jeol 1200 electron microscope (Jeol Co.; Toyko, Japan) in Kuwait. The cell types were identified and photographed.
Nucleocytoplasmic Ratio
This ratio was measured in micrographs of cells including whole nucleus and cytoplasm with the help of a Zeiss MOP-3 (Manual Optical Planimeter, Zeiss; Esslingen, Germany). First, the surface areas of the whole cell and then that of the nucleus were estimated. The nucleocytoplasmic ratio was estimated by dividing the nuclear area by the cell area minus the nuclear surface area.
Organelle Measurement
The sizes of mitochondria and secretory granules, if present, were estimated in six or more electron micrographs per cell type at x15,000 magnification. Thus, the widths of clearly outlined mitochondria were recorded and averaged for each cell type. The average diameter of sharply delineated secretory granules was similarly measured, and, when the granules were not spherical, the longest and shortest axes were recorded and averaged. The results are expressed as means ± standard error (SE). Difference in any two measurements was assessed by the Student’s t test. p values were considered significant between 0.01 and 0.05, and highly significant below 0.01.
|
RESULTS
|
|---|
Examination of the human oxyntic epithelium by the light microscope showed the location of the four mature epithelial cell types: pit cells along the pit, neck cells along the gland neck, zymogenic cells along the gland base, and parietal cells throughout the gland (Fig. 1
). However, there was extensive variation in pit-gland features. For instance, pit and gland depth was highly variable (Table 2), with the gland-to-pit ratio averaging 2.3. In spite of variation, it was possible to define limits between regions. Thus the pit-isthmus boundary was assigned to the high edge of the highest parietal cell, the isthmus-neck boundary to the high edge of the highest neck cell, and the neck-base boundary to the high edge of the highest pre-zymogenic cell (a cell to be defined later). Finally, each region under study, whether pit, isthmus, neck, or base, was arbitrarily divided into three equal segments referred to as high (near gastric lumen), low (far from gastric lumen), and mid (between the other two).
Electron microscopic examination confirmed the cell features listed in Table 1, and revealed the structure of cells described below.
Immature Cells
The immature cells were sorted into four classes, which share a number of features. The first common feature is their location in the isthmus region. The second feature is their size, which is smaller than that of mature cells. The third is a high nucleocytoplasmic ratio, about twice or more that of mature cells (Table 3). The fourth common feature (Figs. 2A–2D) is the presence in the nucleus of mainly diffuse chromatin and prominent reticulated nucleoli. The fifth is the presence of many free ribosomes in the cytoplasm, while other organelles are few and small (Figs. 2A–2D). The last common feature is the small size of their mitochondria with a width estimated at 225–304 nm, which is smaller than in mature cells (Table 4). In spite of their common features, immature cells exhibit sufficient differences to allow their subdivision into four classes; cells in classes 2–4 present unequivocal features, whereas class 1 cells are not defined as precisely.
View larger version (177K):
[in this window]
[in a new window]
|
Figure 2. The four classes of immature cells in the isthmus region of the gland. The cells in panels A-D face the glandular lumen (L) at the top and the basement membrane (BM) at the bottom. (The cells depicted in Figs. 3–7 are in the same orientation.) The cells in the four classes exhibit large nuclei, which contain mainly diffuse chromatin and whose nucleoli, when visible (n), appear reticulated, whereas the scanty cytoplasm includes small mitochondria (m), few cisternae of rough ER, a discrete Golgi apparatus (G), and many free ribosomes. A) Class-1 cells are characterized by the presence in the apex of minute secretory granules with variegated content and, accordingly, named mini-granule cells. The small Golgi apparatus contains a granule (arrow) of the same size as those in the apex. At upper right, an inset showing the apex of a mini-granule cell viewed at higher magnification (x10,600) than the cell itself (x5,500) displays dense (dg) and cored (cg) granules. B) Class-2 cells are characterized by the presence in the apex of a few dense, homogeneous granules. Similar granules in the paranuclear Golgi region vary slightly in size and density. At the cell apical surface, microvilli have a denser glycocalyx than in the other three panels. For reasons indicated in the text, this cell has been named pre-pit cell. Besides this cell, three other cells are visible in the picture. At right, there are two other class 2 cells, with the farther one showing two lobes of the nucleus and a Golgi apparatus, while at left, a parietal cell exhibits many large mitochondria (M), a few lysosomal bodies (Ly) and apical microvilli (MV), x4,100. C) Class 3 cells are characterized by small cored secretory granules (cg). In the uppermost granule, the dark background is limited to a crescent, which is visible at its left and partly encloses the light spherical core. Light- and dark-appearing cored granules are also seen nearby. Such cored granules may be cut tangentially through the background and then appear all dark, or through the core and then appear all light. Similarly varied granules are seen in the Golgi region. Besides this cell, the upper left corner includes part of a neck cell in which cored granules are larger and more numerous than in the class 3 cells. As indicated in the text, the class 3 cell has been named pre-neck cell, x5,500. D) Class 4 cells are characterized by their wide base, relatively long microvilli piled up in the gland lumen, and the presence in the cytoplasm of closed microvillus-containing structure, which, because of its similarity to parietal cell canaliculi, is referred to as small canaliculus (c). Close to it are a few small granules (g), including the dense and cored varieties. The nature of the large, more or less circular vesicles containing frothy material enclosed within a dense membrane (*) has not been identified. Part of a similar cell is seen at lower right. In both cells, the small clear vesicles are cross-sections of a tubulovesicular system similar to that in parietal cells. These are abundant in the apex of the right cell (TV). For reasons made clear in the text, the class 4 cell has been named pre-parietal cell, x5,500.
|
|
Class 1 immature cells are characterized by the presence in the apex of tiny granules varying in density and appearance, but fairly uniform in size (Fig. 2A). Some granules are dense throughout, while others are composed of a light core in a dense background; the light core may be in the center of the granule but is often at its periphery (inset in Fig. 2A). The presence of similar granules on the trans face of the Golgi apparatus and along the way to the surface indicates that the granules observed in the apex are of Golgi origin and are, presumably, secretory. The cells in class 1 are somewhat diverse; moreover, they are infrequent. Thus, only six of them have been observed in the four biopsies examined after their initial identification. The justification of their pooling into a distinct category is the uniformly small size of their secretory granules (mean = 219 nm; Table 5). Class 1 cells are usually located in mid isthmus (Fig. 1). Finally, when these cells were compared with the four mature cell types, they did not show similarity to any of them.
Class 2 immature cells display apical granules that are uniformly homogeneous and dense (Fig. 2B) and have a mean diameter about twice that of mini-granule cells (Table 5). Within the Golgi region, uniformly homogeneous granules varying somewhat in density are taken to be developing secretory granules destined for the apex. These cells have a regular columnar shape; they are numerous and mainly found in high isthmus. The microvilli are rather short but coated by a thicker glycocalyx than in other immature cells. Mitoses of these cells have been observed. Finally, when class 2 immature cells are compared with the mature cell types (pit cells), their secretory granules, even though few and small, display the same density and homogeneity (Fig. 3A, 3B). Moreover, in class 2 cells, as in pit cells, microvilli have a dense glycocalyx. Finally, the high isthmus segment is a neighbor to the low pit segment, so that some of the class 2 cells in the former come in contact with some of the pit cells in the latter. In fact, the successive cells at the boundary between the two segments display a gradual transition in granule size and number (Fig. 1). For these various reasons, we have postulated that class 2 cells are the progenitors of pit cells. They have been tentatively designated as pre-pit cells.
View larger version (166K):
[in this window]
[in a new window]
|
Figure 3. A) Apical region of a low pit cell located close to the pit-isthmus border. This cell has dense secretory granules (dg) similar to, but larger and more numerous than, those in pre-pit cells, thus indicative of a differentiating pit cell. The granules are scattered throughout the apex, where they are interspersed with mitochondria (m), rough ER cisternae and Golgi apparatus (G). Part of a parietal cell at upper right shows large mitochondria (M), x7,600. B) Two pit cells located at about the junction of the mid and high pit segments. A characteristic feature is the packing of the numerous apical dense granules within an organelle-free region of the cell referred to as ectoplasm. The main organelles, mitochondria, Golgi, and ER cisternae are found outside the pack. Along the basal region of the cells, wide intercellular spaces swarm with interdigitating cytoplasmic processes, an observation common in the cells at this level of the pits, x4,500. C) Apical region in a high neck cell located close to the isthmus-neck border. The cell carries cored granules (cg), which are similar to, but larger and more numerous than, those in pre-neck cells. Even though such granules generally have a light core in a dense background, this particular figure shows an unusual number of granules that have the reverse pattern of a dense core in a light background (arrow). A small portion of the nucleus is seen in the lower left corner, x12,100. D) Two mid neck cells. They are characterized by numerous apical cored granules. The granules share the supranuclear cytoplasm with Golgi apparatus, rough ER cisternae, and mitochondria. The neck cell at the right has larger granules than the other and appears to be at a more advanced state of differentiation. At lower left and upper right are parietal cells exhibiting large mitochondria, x7,600.
|
|
Class 3 immature cells are characterized by the presence of a few apical, 430-nm wide, bipartite granules composed of a dense background within which appears a light, spherical core (Fig. 2C). The presence of cored granules in the Golgi region indicates that those found in the apex are secretory granules. In contrast to the typical cored granules described, an occasional one shows a reverse pattern, that is, a dense core in a light background. The number of granules in class 3 cells is usually small (Fig. 2C). The presence of mitoses of these cells has been observed. Finally, when class 3 cells are compared with the four mature cell types, only the neck cell displays cored granules (Figs. 3C, 3D, 4A). Also, the low isthmus segment in which most class 3 cells are located is a neighbor to the high neck that harbors neck cells. While pre-neck and neck cells may be side by side (Fig. 2C), the usual pattern is a progressive increase in the granule number of the successive cells at the isthmus-neck boundary (Fig. 1). These various reasons have led us to postulate that class 3 cells are the progenitors of neck cells. They have been tentatively given the name pre-neck cells.
View larger version (169K):
[in this window]
[in a new window]
|
Figure 4. Cells found about the junction between gland neck and base. A) Low neck cell exhibiting numerous cored granules in the supranuclear region, x5,900. B) Two cells whose granules combine features of neck and zymogenic cell secretory granules. Most of the granules are bipartite since they contain darkly stained mucus and lightly stained material presumed to be rich in pepsinogen. In the cell at left, the dark mucus predominates, whereas in the cell at right, the light material prevails. These two cells are designated respectively subtype-1 pre-zymogenic cell and subtype-2 pre-zymogenic cell, x 7,500. C) The cell in this panel also has granules combining features of neck and zymogenic cell granules, but the light pepsinogen-rich material greatly exceeds the dark mucus in most granules. This cell is designated a subtype-3 pre-zymogenic cell. Golgi stacks and mitochondria are dispersed between the granules. Rough ER cisternae accumulate between cell base and nucleus. At upper left, there is a parietal cell exhibiting large mitochondria and a canaliculus (C), x7,500. D) Zymogenic cell characterized by large apical secretory granules whose homogeneously light content is known to be rich in pepsinogen. Mitochondria, Golgi stacks, and single rough ER cisternae are scattered among the granules, while stacked rough ER cisternae occupy the base of the cell, x9,500.
|
|
Class 4 immature cells differ from the others by a squat shape (Fig. 2D), relatively long microvilli, and the presence of few tubules and vesicles of equal diameter in the cytoplasm (Fig. 5A). The nucleus is oriented parallel to the basement membrane, whereas it is perpendicular in the other immature cells. Microvilli are longer than in other immature cells but their glycocalyx is thin. In the apical cytoplasm, the main feature is a rather circular, membrane-bound structure referred to as small canaliculus, which, like the large canaliculi observed in parietal cells, encloses microvilli. The small canaliculus may be entirely within the cytoplasm (Fig. 2D), connected to the surface (Fig. 5B), or missing (Fig. 5A). The Golgi apparatus is small. A small number of variously sized cored and dense granules can be found in the cytoplasm (Fig. 2D). Class 4 cells are distributed through the various segments of the isthmus, where they are less numerous than pre-pit and pre-neck cells but more so than mini-granule cells. Mitosis of these cells has not been observed. Finally, when class 4 cells are compared with mature cells, it is clear that they may include canaliculi, vesicles and tubules, and relatively long microvilli, all attributes of parietal cells (Table 1). Hence, we have postulated that class 4 cells are the progenitors of parietal cells and have been tentatively named pre-parietal cells.
View larger version (103K):
[in this window]
[in a new window]
|
Figure 5. Separating pre-parietal cells into three subtypes according to the degree of differentiation. A) The least differentiated pre-parietal cell is characterized by the absence of small canaliculi, even though other typical features of pre-parietal cells, namely, scattered tubulovesicles (TV) and long apical microvilli (MV) coated with a relatively thin glycocalyx are present. Such poorly differentiated cells are referred to as pre-parietal-1 cells, x 9,500. L = glandular lumen. B) A more differentiated subtype is characterized by the presence of small canaliculi. One is seen here at lower right, while another at upper center appears as an invagination of the apical plasma membrane. In addition, there are tubulovesicles and long microvilli. Such cell is typical of pre-parietal-2 cells, x 7,500. MB = multivesicular body. C) The most differentiated pre-parietal cell is large. Here, besides two sections of canaliculi, tubulovesicals are more numerous, mitochondria are bigger, and chromatin is denser than in Figure 5B. Such cells are referred to as pre-parietal-3 cells, x 9,500.
|
|
Associations of Immature and Mature Cells
Pre-Pit and Pit Cells
The light microscope showed not only the gradual transition mentioned above between high isthmus pre-pit cells and low pit cells, but also a continuation of progressive changes along the pit. Toward the mid pit, granules were gradually packed in the cell apex (Fig. 3B), while up to the top of the high pit, the number of granules per pack progressively increased (Fig. 1). The sequence could indicate that pre-pit cells are the source of the progressively changing pit cells. Accordingly, we proposed that pre-pit cells are the progenitors of the whole pit cell lineage.
Pre-Neck, Neck, and Zymogenic Cells
A gradual transition was observed not only between pre-neck and neck cells at the isthmus-neck boundary, as seen in Figure 1, but also between neck and zymogenic cells at the neck-base boundary, as seen by changes in their secretory granules detected by the electron microscope. At this boundary, the granules present in some of the cells defined the transition between neck and zymogenic cells by their intermediate size (Table 5) and by their shape and content, whose variation allowed the identification of three subtypes of these cells. In the first (Fig. 4B, left), the light core and dark content of most granules are about equal in size. In the second subtype (Fig. 4B, right), the pale core is larger than the dense component, which appears ragged, sometimes faintly stippled, and either centrally or peripherally positioned. In the third subtype, the electron-dense component of the granules is minimal at the periphery, often reduced to a small cap or crescent (Fig. 4C), while the light component is large, so that some of the granules resemble those in zymogenic cells. Finally, for comparison, zymogenic cells are depicted in Figure 4D with spherical, large secretory granules whose content is entirely light. Since the features of the three subtypes are intermediate between those of neck and zymogenic cells, they are respectively referred to as pre-zymogenic-1, -2, and -3 cells. The sequence from pre-neck cells through neck and pre-zymogenic cells to zymogenic cells is characterized by a progressive increase in the size of secretory granules (Table 5) and mitochondrial width (Table 4). The sequence could indicate that neck cells give rise not only to pre-zymogenic cells but also, through them, to zymogenic cells. Accordingly, we proposed that pre-neck cells are the progenitors of the lineage ending in zymogenic cells, that is, the zymogenic cell lineage.
Pre-Parietal and Parietal Cells
The features of pre-parietal cells (Fig. 2D) are not only similar to those of parietal cells but also different enough to allow their subdivision into three subtypes, namely, pre-parietal-1 cells, which exhibit tubulovesicles but lack canaliculi (Fig. 5A); pre-parietal-2 cells characterized by one or more small canaliculi (Fig. 2D and 5B); and pre-parietal-3 cells, which are somewhat larger and have larger canaliculi than the previous one (Fig. 5C). For comparison, Figure 6 presents a mature parietal cell in which canaliculi and tubulovesicles are extensive, while mitochondria are large and numerous even though their size increases from isthmus to base (Table 4, last three lines). When pre-parietal subtypes are compared with one another and to parietal cells, it was noted that size, microvillus length, and mitochondrial width (Table 4) gradually increase from pre-parietal-1 cells to parietal cells, thus forming a series that could represent stages in the development of parietal cells. Accordingly, we proposed that pre-parietal-1 cells are the progenitors of the parietal cell lineage.
View larger version (210K):
[in this window]
[in a new window]
|
Figure 6. Parietal cell exhibiting several profiles of secretory canaliculi, numerous elements of the tubulovesicular system, and many large mitochondria. Long microvilli project from the apical and canalicular membranes, x5,200.
|
|
Degenerating Cells
The uppermost pit cells and those at the luminal surface may display changes, such as pyknosis of the nucleus (depicted as a dark spot in the epithelium at upper left in Fig. 1), disintegration of secretory granules (depicted as darkened granules in a cell at upper right in Fig. 1), decrease in mitochondrial size (left of Fig. 7A), and occasional massive cell degeneration with variously sized vacuoles and lysosomal bodies (center of Fig. 7A).
View larger version (146K):
[in this window]
[in a new window]
|
Figure 7. Degenerating cells. A) Degenerating pit cell. Along the surface of the oxyntic mucosa facing the gastric lumen, the figure shows from left to right: thin segments of pit cell cytoplasm, one of which includes the small mitochondria found at this stage of pit cell life, and the degenerating pit cell that occupies most of the figure. At lower right, its nucleus shows chromatin beginning to condense into small dense dots. In the rest of the cell, vacuoles are visible, the most prominent of which (V) includes irregular debris. Some secretory granules start degenerating (g). Lysosomes (Ly) are present, x6,400. B) Degenerating parietal cell. Clockwise from the upper right corner is a neck cell, the degenerating parietal cell in which abnormal canaliculi and mitochondria are separated by multiform vacuolar spaces, then another neck cell, and finally a normal parietal cell showing tubulovesicles and long microvilli protruding into the glandular lumen, x5,500. C) Degenerating zymogenic cell. Down from the upper left corner is a normal zymogenic cell, the degenerating zymogenic cell in which condensation of the nuclear chromatin is advanced and distension of ER cisternae, including the perinuclear cisterna, is pronounced, and finally another degenerating zymogenic cell, x 5,000.
|
|
In the neck, and more frequently in the base region, an occasional parietal cell undergoes degeneration characterized by extensive dissolution of canaliculi and intermitochondrial cytoplasm with the appearance of diverse vacuoles (Fig. 7B).
In the low base, the odd zymogenic cell undergoes degeneration, as indicated by condensed nucleus and distended rough endoplasmic reticulum (ER) cisternae (Fig. 7C). Degenerating neck cells have not been observed.
|
DISCUSSION
|
|---|
Our study of the human oxyntic epithelium led to three proposals on the respective role played by the immature class 2 (pre-pit), 3 (pre-neck), and 4 (pre-parietal) cells as progenitors of the pit, zymogenic, and parietal cell lineages, respectively. However, though our morphological observations were highly suggestive, they did not provide decisive evidence. In the hope of finding further support for the proposals, advantage was taken of observations on the mouse stomach completed with the help of timed 3H-thymidine radioautography, and demonstrated, in particular, how immature cells were involved in the production of mature cells [9–11]. We, therefore, decided to systematically compare the present observations in the human with the observations clarified by radioautography in the mouse.
To accomplish this aim, the role of the immature cells as progenitors will be discussed by comparing the three groups of cells analyzed in the human with the relevant cells in the mouse, that is, first the immature cells themselves, then the associations of these cells with mature cells, and finally the degenerating cells. This discussion is completed by remarks on the source of the immature cells and on the implications of the results for the future.
Immature Cells
When the four human immature cell classes (Figs. 2A–2D) were compared with relevant mouse cells (Figs. 11–14 in [26]), the striking finding was that three of them (pre-pit, pre-neck, and pre-parietal cells) were closely similar in the two species, a conclusion justifying their being given the same name in both. In contrast, the fourth immature cell in each species differed entirely from that in the other. Thus, in the mouse, the common granule-free cell (Fig. 11 in [26]), a nonsecretory cell identified as the multipotent stem cell of the epithelium, had no counterpart in the human. In contrast, the occasionally observed mini-granule cell in the human, with apical granules indicative of secretory activity, had no equivalent in the mouse.
Associations of Immature and Mature Cells
Pit Cell Lineage
When the pit-isthmus boundary in the human was compared with that of the mouse (Fig. 8), the sequence of pre-pit cells in the isthmus and pit cells along the pit appeared similar in the two species, so the cell behavior was likely to be similar, too. The use of timed 3H-thymidine radioautography in the mouse revealed that pre-pit cell mitoses yielded cells that entered the pit where, as pit cells, they ascended to the surface [9]. Mitoses were observed in human pre-pit cells. It was inferred from the similarity of the human structure to that of the mouse that, in the human, mitoses of the pre-pit cell were the source of pit cells and these ascended the pit. This reasoning was in accord with the report cited in the Introduction that cultured human gastric biopsies exposed to labeled thymidine showed migration of pit cells to the surface [12, 13, 16]. Hence, comparison of the cell association at the human and mouse pit-isthmus boundary provided evidence in support of the first of the three proposals, namely, pre-pit cells are the progenitors of the pit cell lineage in humans.
View larger version (34K):
[in this window]
[in a new window]
|
Figure 8. Mouse oxyntic mucosa drawn in the light microscope. The pit is depicted in continuity with a single gland composed of isthmus, neck, and base (but the pit could be in continuity with two or more glands). The pit is lined by pit cells. The apex of these cells includes dense granules, whose number per cell progressively increases from the low pit to the surface, where it is sharply reduced as cells face the gland lumen. The gland isthmus is mainly composed of four types of immature cells, namely, granule-free cells, pre-pit, pre-neck, and pre-parietal cells. The gland neck is lined by neck cells and the gland base by zymogenic cells. Parietal cells and the odd enteroendocrine cell are scattered throughout. Finally, the following cells can be in close contact: pre-pit and pit cells at the pit-isthmus boundary; pre-neck and neck cells at the isthmus-neck boundary; neck, pre-zymogenic, and zymogenic cells at the neck-base boundary (drawn by Margo Oeltzschner). Reproduced with permission from [8].
|
|
Zymogenic Cell Lineage
When the isthmus-neck boundary was compared in humans (Fig. 1) and mice (Fig. 8), the sequence from pre-neck to low neck cells appeared similar in both. Since the use of timed 3H-thymidine radioautography in the mouse had shown that the label was transferred from pre-neck to neck cells while they were migrating downward [10], the similarity of the sequence in the two species supported the conclusion that, in humans as in mice, pre-neck cells produced neck cells while both were migrating downward. However, when the neck-base boundary was compared in Figures 1 and 8, the tight grouping of the cells prevented effective light microscopic comparison as done above. On the other hand, the electron microscope revealed that, at this site, the three cell types designated pre-zymogenic-1, -2, and -3 cells displayed granules showing features gradually changing from those of neck cells to those of zymogenic cells. The three pre-zymogenic subtypes in the human (respectively, Figs. 4B left, 4B right, and 4C) were quite similar to those of the mouse (respectively, Figs. 7, 8 right, and 8 center in [10]). In the mouse, timed 3H-thymidine radioautography demonstrated passage of label from cell to cell, indicating not only that pre-neck cells transformed into neck cells, as mentioned above, but also that neck cells eventually changed into pre-zymogenic cells, which then became zymogenic cells. From the close structural similarity, it was inferred that, in humans too, neck cells transformed through intermediate pre-zymogenic stages into zymogenic cells. The successive evolution of pre-neck cells into neck, pre-zymogenic, and zymogenic cells received indirect support from the cell-to-cell increase in the size of secretory granules (Table 5) and mitochondria (Table 4). In the past, the conclusion that neck cells transformed into zymogenic cells in humans had been reached by Cornaggia et al. [22, 23]. Hence, the comparison of the human and mouse cell associations at the isthmus-neck and neck-base boundaries, as well as the increase in size of granules and mitochondria, provided evidence in support of the second proposal, namely, pre-neck cells were the progenitors of the zymogenic cell lineage in humans.
Parietal Cell Lineage
When human pre-parietal-1, -2, and -3 cells (respectively, Figs. 5A, 5B, and 5C) were compared with those in the mouse (respectively, Fig. 14 in [26], Figs. 7 and 8 in [11]), it was noted that in both species, pre-parietal-1 cells displayed fairly long microvilli and a few tubulovesicles and pre-parietal-2 cells were characterized by small canaliculi, while pre-parietal-3 cells were larger in size and had more canaliculi than the others. Finally, mature parietal cells in the human (Fig. 6) and mouse (Fig. 3 in [26]) displayed the fully developed listed features. Hence, the sequence from pre-parietal-1 to parietal cell was quite similar in human and mouse. In the mouse, the sequence was shown to reflect the steps in the development of the lineage, even though the thymidine label had not been traced by radioautography from pre-parietal-1 to pre-parietal-3 cells because of their low number. Yet the passage of radiolabel from pre-parietal to parietal cells was demonstrated, and radioautography was utilized to tag the migration of parietal cells out of the isthmus in either of two directions, that is, outward to the pit or downward to the neck and base [11]. In the human, the three pre-parietal cells, as well as some parietal cells, were present in the isthmus as in the mouse, suggesting that, in both species, pre-parietal-1 cells gave rise to the sequence ending in parietal cells at this site. However, the absence of parietal cells in the pit and their presence in neck and base indicated that they migrated only downward. Hence, the evidence presented was in accord with the third proposal, namely, pre-parietal-1 cells were the progenitors of the parietal cell lineage in humans.
Degenerating Cells and Lineage Renewal
In the pit cell lineage of the mouse, pit cells present at the surface produced few small granules, resulting in a decreased size of the granule pack size (Fig. 8), while some cells underwent degeneration and others were lost to the lumen [9]. In the human, too, when pit cells reached the surface, they showed a decrease in the granule pack (Fig. 1) as well as massive degeneration in some cells (Fig. 7A). Past work mentioned in the Introduction indicated that the migration of human pit cells from the pit bottom up to the free surface took about three days [14, 15], where 2.4% of them died [17], while the others desquamated at the rate of 5.5 x 105 cells per minute [18]. Maintenance of the steady state of the epithelium required that such rapid rate of cell death be balanced by an equally rapid rate of cell production. Since we observed mitoses of pre-pit cells but not of pit cells, it was likely that the rapid cell production was essentially due to high mitotic activity of pre-pit cells. It was concluded that the cells in the human pit cell lineage were being rapidly replaced by new cells, probably in a continuous manner. Hence this lineage was considered to be a renewal system.
In the zymogenic cell lineage, only the mature zymogenic cells, ending the line of descent from pre-neck cells through neck and pre-zymogenic cells, showed occasional signs of advanced degeneration (Fig. 7C). Thus, zymogenic cell death was taken to be the terminal event in the cell lineage. Degenerating zymogenic cells were rarely seen. To maintain the steady state, cell production, attributed mainly to pre-neck cell mitoses, must also occur at a relatively low rate, and so must the overall turnover rate. Hence the zymogenic cell lineage was taken to be a renewal system turning over at a relatively low rate.
In the parietal cell lineage, the proportion of degenerating parietal cells in the mouse was 21% and 23% within pit and base, respectively [11]. In the human, parietal cells migrated only toward the base, where signs of their advanced degeneration were observed (Fig. 7B). Hence the death of parietal cells in the base was likely to be the terminal event in the human lineage. This degeneration was fairly common and, presumably, was taking place at a fair rate. Maintenance of the steady state required that the rate of pre-parietal cell production (by a mechanism conjectured below) occurred at an equally fair rate. Hence, the parietal cell lineage was considered to be a renewal system turning over at a fair rate.
In conclusion, the comparison of human and mouse oxyntic epithelium revealed that each one of the three immature cells, pre-pit, pre-neck, and pre-parietal cells, were progenitors giving rise to a distinct cell lineage, in which cells underwent renewal. The implication was that, at least in human and mouse, the cells in the three lineages, that is, practically all the cells in the epithelium, were being replaced by new ones, presumably in a continuous manner. If, by any chance, any cell of the oxyntic epithelium was damaged by food, local infection, or toxin attack, this cell would soon be eliminated like any other. Such elimination of damaged cells is likely to have survival value. Hence, cell renewal in the three lineages was a beneficial process, which, in the course of evolution, could have been one of the factors allowing for survival of the species.
Source of Immature Progenitor Cells
In the mouse, the three progenitors, pre-pit, pre-neck, and pre-parietal cells, were found to come from the multipotent granule-free cell [27]. In the human, such a cell was not found and the origin of the progenitor cells was not elucidated. Nevertheless, two possible hypotheses will be considered.
Multipotentiality Hypothesis
The fact that no multipotential cell was found in the human does not eliminate the possibility that one could be discovered in the future. Furthermore, even though there was no direct evidence that the mini-granule cell played a multipotent role, this remained a possibility since this cell contained small dense, homogeneous granules (suggesting possible development toward pre-pit cells), as well as small cored granules (suggesting possible development toward pre-neck cells), and could be shaped like a pre-parietal cell (indicating the possibility of developing in that direction). Briefly, the hypothesis was that the cells of the epithelium were derived from a multipotent stem cell, but whether this stem cell was the mini-granule cell or some unidentified cell could only be surmised
Self-Generating Hypothesis
Another possibility was that the progenitors regenerated themselves by mitosis. We indeed observed mitoses of pre-pit and pre-neck cells, although not of pre-parietal cells. Rubin et al. [1] also observed mitoses, which they depicted in cells that we interpreted as pre-pit (Fig. 18 in [1]) and pre-neck cells (Fig. 14 in [1]). Perhaps pre-pit and pre-neck cells could give rise not only to their own lineage but also to pre-parietal cells. Some support for this hypothesis resides in the fact that pre-parietal cells often include small numbers of dense or cored granules (Figs. 2D and 5), which would have been retained from the hypothetical pre-pit or pre-neck cell precursor. Accordingly, our proposal was that pre-parietal cells were derived from pre-pit or pre-neck cells. Briefly, the hypothesis was that mitosis of pre-pit and pre-neck cells produced cells for renewal of the three main cell lineages (Fig. 9).
View larger version (39K):
[in this window]
[in a new window]
|
Figure 9. Model of cell differentiation in the epithelium of the human oxyntic mucosa, based on the self-generating hypothesis. This hypothesis is that mitoses of two progenitors, pre-pit and pre-neck cells (drawn at left), supply cells for the whole epithelium. Thus mitoses of pre-pit cells not only insure their own renewal, but also, as shown by the top line, give rise to pit cells that achieve their differentiation to the mature state while ascending the pit, thus completing the pit cell lineage. Below, mitoses of pre-neck cells not only insure their own renewal, but also, as shown by the bottom line, give rise to neck cells that transform through a pre-zymogenic step into zymogenic cells, thus completing the zymogenic cell lineage. Finally, in the middle, as indicated by broken lines, pre-pit and pre-neck cells also give rise to pre-parietal-1 cells which then differentiate into the sequence of pre-parietal-2, pre-parietal-3 and eventually mature parietal cells, thus completeing the parietal cell lineage.
|
|
|
IMPLICATIONS FOR THE FUTURE
|
|---|
In many respects, the human oxyntic epithelium resembles that of the mouse. Hence, the recent discoveries on the control of cell dynamics in mouse stomach could be relevant to the human. In general, hormones such as gastrin [28], as well as peptides such as the epidermal growth factor and transforming growth factor alpha and beta, played a role in controlling cell proliferation and differentiation in the mouse oxyntic epithelium [29]. Genetic manipulations resulting in overexpression of transforming growth factor alpha [30] or activins [31], or involving more complex combinations [32, 33], concurred in indicating that a critical step in the control of the epithelium was the production of parietal cells from pre-parietal cells. The inhibition of parietal cell formation blocked the development of zymogenic cells, while leading to an amplification of immature cells [32, 33]. In the future, methods could be developed to find out whether the control in the human oxyntic epithelium resided in factors associated with the production of parietal cells from pre-parietal cells, as appeared to be the case in the mouse. Hence, while the three lineages identified in the human oxyntic epithelium appeared to be independent, they could be interrelated since, in the mouse, the parietal cell lineage influenced the others.
|
ACKNOWLEDGMENT
|
|---|
This work was carried out with the support of grants from Kuwait University and Terry Fox Foundation (to S.M.K.) and the Medical Research Council of Canada (to C.P.L.). The authors are grateful to Dr. Gary Wild and Dr. Dia Mamoon for providing the biopsies, to Yves Clermont for drawing Figure 1, and to Gerald Buzzell for his comments on the manuscript.
|
REFERENCES
|
|---|
- Rubin W, Ross LL, Sleisenger MH et al. The normal human gastric epithelia. A fine structural study. Lab Invest 1968;19:598–626.[Medline]
- Helander H, Leth R, Olbe L. Stereological investigations on human gastric mucosa: I. normal oxyntic mucosa. Anat Rec 1986;216:373–380.[CrossRef][Medline]
- Owen DA. Stomach. In: Sterenberg SS, ed. Histology for Pathologists. New York: Raven, 1992:533–545.
- Fawcett DW. In: Jensh RP, ed. Bloom and Fawcett: a Textbook of Histology. New York: Chapman & Hall, 1994:599–616.
- Corpron RE. The ultrastructure of the gastric mucosa in normal and hypophysectomized rats. Am J Anat 1966;118:53–90.[CrossRef][Medline]
- Wattel W, Geuze JJ, de Rooij DG. Ultrastructural and carbohydrate histochemical studies on the differentiation and renewal of mucous cells in the rat gastric fundus. Cell Tissue Res 1977;176:445–462.[Medline]
- Kataoka K. Electron microscopic observations on cell proliferation and differentiation in the gastric mucosa of the mouse. Arch Hist Jpn 1984;47:209–221.
- Karam SM, Leblond CP. Dynamics of epithelial cells in the corpus of the mouse stomach. I. Identification of proliferative cells and pinpointing of the stem cell. Anat Rec 1993;236:259–279.[CrossRef][Medline]
- Karam SM, Leblond CP. Dynamics of epithelial cells in the corpus of the mouse stomach. II. Outward migration of pit cells. Anat Rec 1993;236:280–296.[CrossRef][Medline]
- Karam SM, Leblond CP. Dynamics of epithelial cells in the corpus of the mouse stomach. III. Inward migration of neck cells followed by progressive transformation into zymogenic cells. Anat Rec 1993;236:297–313.[CrossRef][Medline]
- Karam SM. Dynamics of epithelial cells in the corpus of the mouse stomach. IV. Bidirectional migration of parietal cells ending in their gradual degeneration and loss. Anat Rec 1993;236:314–332.[CrossRef][Medline]
- Tanaka J. Autoradiographic studies of the cell proliferation of the human gastric mucosa in supravital condition. Acta Pathol Jpn 1968;18:307–318.[Medline]
- Hansen OH, Pedersen T, Larsen JK. Cell proliferation kinetics in normal human gastric mucosa. Gastroenterology 1976;70:1051–1054.[Medline]
- Castrup HJ, Fuchs K. [Cell renewal in inflammatory changes of the gastric mucosa]. Dtsch Med Wochenschr 1974;99:892–895.[Medline]
- Hart-Hansen O, Johansen AA, Larsen JK et al. Cell proliferation in normal and diseased gastric mucosa. Acta Path Microb Scand 1979;87:217–222.
- Helander HF. Kinetics of the gastric epithelial cells. In: Domschke W, Konturek S, eds. The Stomach: Physiology, Pathophysiology and Treatment. Berlin: Springer-Verlag, 1993.
- Mackercher PA, Ivey KJ, Baskin WN et al. A scanning electron microscopic study of normal human oxyntic mucosa using blunt dissection and freeze fracture. Am J Dig Dis 1978;23:449–459.[Medline]
- Croft DN, Pollock DJ, Coghill NF. Cell loss from human gastric mucosa measured by the estimation of deoxyribonucleic acid (DNA) in gastric washings. Gut 1966;7:333–343.[Free Full Text]
- Eastwood GL. Gastrointestinal epithelial renewal. Gastroenterology 1977;72:962–975.[Medline]
- Lipkin M. Growth and development of gastrointestinal cells. Annu Rev Physiol 1985;47:175–197.[CrossRef][Medline]
- Plenk H. Der Magen. In: Von Moellendorff’s Handbuch der mikroscopischen Anatomie des Menschen. vol. 5, part 2. Von Moellendorff W, ed. Berlin: Springer-Verlag, 1932.
- Cornaggia M, Capella C, Riva C et al. Electron immunocytochemical localization of pepsinogen I in chief cells, mucous neck cells and transitional mucous neck/chief cells of the human fundic mucosa. Histochemistry 1986;85:5–11.[CrossRef][Medline]
- Cornaggia M, Riva C, Capella C et al. Subcellular localization of pepsinogen II in stomach and duodenum by the immunogold technique. Gastroenterology 1987;92:585–593.[Medline]
- Waalewijn RA, Meuwissen SG, Pals G et al. Localization of pepsinogen (A and C) and cellular differentiation of pepsinogen-synthesizing cells in the human gastric mucosa. Eur J Cell Biol 1991;54:55–60.[Medline]
- Hanby AM, Poulsom R, Playford RJ et al. The mucous neck cell in the human gastric corpus: a distinctive, functional cell lineage. J Pathol 1999;187:331–337.[CrossRef][Medline]
- Karam SM, Leblond CP. Identifying and counting epithelial cell types in the "corpus" of the mouse stomach. Anat Rec 1992;232:231–246.[CrossRef][Medline]
- Karam SM, Leblond CP. Dynamics of epithelial cells in the corpus of the mouse stomach. V. Behavior of entero-endocrine and caveolated cells: general conclusions on cell kinetics in the oxyntic epithelium. Anat Rec 1993;236:333–340.[CrossRef][Medline]
- Walsh JH. Gastrointestinal hormones. In: Johnson LR, ed. Physiology of the Gastrointestinal Tract. Vol 1, 3rd ed. New York: Raven, 1994:1–128.
- Podolsky DK. Peptide growth factors in the gastrointestinal tract. In: Johnson LR, ed. Physiology of the Gastrointestinal Tract. Vol 1, 3rd ed. New York: Raven, 1994:129–167.
- Sharp R, Babyatsky MV, Takagi H et al. Transforming growth factor alpha disrupts the normal program of cellular differentiation in the gastric mucosa of transgenic mice. Development 1995;121:149–161.[Abstract]
- Li Q, Karam SM, Coerver CA at al. Stimulation of activin receptor II signaling pathways inhibits differentiation of multiple gastric lineages. Mol Endocrinol 1998;12:181–192.[Abstract/Free Full Text]
- Li Q, Karam SM, Gordon JI. Simian virus 40 T antigen amplification of pre-parietal cells in transgenic mice: effects on other gastric epithelial cell lineages and evidence for a p53-independent apoptotic mechanism that operates in a committed progenitor. J Biol Chem 1995;270:15777–15788.[Abstract/Free Full Text]
- Li Q, Karam SM, Gordon JI. Diphteria-toxin mediated ablation of parietal cells in the stomach of transgenic mice. J Biol Chem 1996;271:3671–3676.[Abstract/Free Full Text]
- Lillibridge CB. The fine structure of the human gastric mucosa. Gastroenterology 1964;47:269–296.[Medline]
- Hogben CA, Kent TH, Woodward PA et al. Quantitative histology of the gastric mucosa of man, dog, guinea pig and frog. Gut 1974;67:1143–1154.
Received October 17, 2002;
accepted for publication February 28, 2003.
This article has been cited by other articles:
|
|
|
|
 
D. Heitzmann and R. Warth
No Potassium, No Acid: K+ Channels and Gastric Acid Secretion
Physiology,
October 1, 2007;
22(5):
335 - 341.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
V. G. Ramsey, J. M. Doherty, C. C. Chen, T. S. Stappenbeck, S. F. Konieczny, and J. C. Mills
The maturation of mucus-secreting gastric epithelial progenitors into digestive-enzyme secreting zymogenic cells requires Mist1
Development,
January 1, 2007;
134(1):
211 - 222.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
S. Hata, S. Koyama, H. Kawahara, N. Doi, T. Maeda, N. Toyama-Sorimachi, K. Abe, K. Suzuki, and H. Sorimachi
Stomach-specific Calpain, nCL-2, Localizes in Mucus Cells and Proteolyzes the beta-Subunit of Coatomer Complex, beta-COP
J. Biol. Chem.,
April 21, 2006;
281(16):
11214 - 11224.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
