Sage Journals: Discover world-class research

Abstract

The rat stomach is rich in endocrine cells. The acid-producing (oxyntic) mucosa contains ECL cells, A-like cells, and somatostatin (D) cells, and the antrum harbours gastrin (G) cells, enterochromaffin (EC) cells and D cells. Although chromogranin A (CgA) occurs in all these cells, its processing appears to differ from one cell type to another. Eleven antisera generated to different regions of rat CgA, two antisera generated to a human (h) CgA sequences, and one to a bovine (b) CgA sequence, respectively, were employed together with antisera directed towards cell-specific markers such as gastrin (G cells), serotonin (EC cells), histidine decarboxylase (ECL cells) and somatostatin (D cells) to characterize the expression of CgA and CgA-derived peptides in the various endocrine cell populations of the rat stomach. In the oxyntic mucosa, antisera raised against CgA_291–319 and CGA_316–321 immunostained D cells exclusively, whereas antisera raised against bCgA_82–91 and CgA_121–128 immunostained A-like cells and D cells. Antisera raised against CgA_318–349 and CgA_437–448 immunostained ECL cells and A-like cells, but not D cells. In the antrum, antisera against CgA_291–319 immunostained D cells, and antisera against CgA_351–356 immunostained G cells. Our observations suggest that each individual endocrine cell type in the rat stomach generates a unique mixture of CgA-derived peptides, probably reflecting cell-specific differences in the post-translational processing of CgA and its peptide products. A panel of antisera that recognize specific domains of CgA may help to identify individual endocrine cell populations.

Keywords

chromogranin A ECL cell D cell A-like cell EC cell G cell processing

Chromogranins occur in secretory granules of most neuroendocrine cells, together with amines and/or regulatory peptides (O'Connor 1983; O'Connor et al. 1983; O'Connor and Frigon 1984; Facer et al. 1985; Fischer-Colbrie et al. 1985, 1987; Varndell et al. 1985; Rindi et al. 1986; Grube et al. 1989; Helle 1990). Rat chromogranin A (CgA) is composed of 448 amino acids with 10 pairs of basic amino acid residues representing potential cleavage sites (Hutton et al. 1988; Iacangelo et al. 1988). Different patterns of CgA proteolysis have been reported in adrenal chromaffin cells, in endocrine cell populations of the pituitary gland, and in endocrine cells of the gastroentero–pancreatic tract (Curry et al. 1991; Barbosa et al. 1991; Dockray et al. 1991; Watkinson et al. 1991; Arden et al. 1994). However, differences in expression of CgA-derived peptides in the gastrointestinal tract have not been systematically correlated to individual endocrine cell populations. In the rat stomach, endocrine cells account for approximately 2% of the total number of epithelial cells and the endocrine stomach is therefore comparable in size to the endocrine pancreas. The rat stomach contains at least seven endocrine cell types that can be distinguished by their histochemical and ultrastructural features (Solcia et al. 1975; Capella et al. 1991; Sundler and Håkanson 1991). The acid-secreting (oxyntic) mucosa contains ECL cells (65–70% of the endocrine cell population), A-like cells (20–25%), somatostatin (D) cells (10%), and a small number of D1/P cells (<1%). The antral mucosa contains roughly equal numbers of gastrin (G) cells, enterochromaffin (EC) cells, and D cells (Solcia et al. 1975; Ekelund et al. 1985).

The ECL cells reside within the basal zone of the oxyntic mucosa and are characterized by their expression of histidine decarboxylase (HDC) and histamine (Håkanson et al. 1986, 1994). Histamine, released from the ECL cells in response to gastrin, is a mediator of acid secretion from the parietal cells (Sandvik and Waldum 1991; Andersson et al. 1996). The function of the A-like cells is unknown. Because no specific histochemical marker for the A-like cells has been reported, their identification is based on ultrastructural criteria (Solcia et al. 1975; Capella et al. 1991; Sundler and Håkanson 1991). D cells are located in both the oxyntic and antral mucosa, whereas G cells and EC cells occur in the antral mucosa only. These cell types are characterized by their content of somatostatin, gastrin, and serotonin, respectively.

The objective of this investigation was to use immunohistochemistry to examine the expression and proteolytic processing of CgA in the different endocrine cell populations of the stomach by employing antisera directed to specific domains of CgA and CgA-derived peptides.

Materials and Methods

Antisera Production

All antisera were raised against rat CgA sequences unless otherwise stated (Figure 1). The generation of eleven of the antisera has been described previously (for references see Table 1). One of these antisera (R880) was raised against a bovine (b) rather than a rat CgA sequence (bCgA82–91). This sequence differs from the corresponding rat sequence in two positions (F in rCgA, instead of Y, in position 82 and F in rat CgA, instead of L, in position 90). Antiserum (R814) was raised against a human (h) CgA sequence (343–356) that differs from the corresponding rat sequence in one position only (R in rCgA instead of K in position 345). An additional three CgA antisera were generated and used here for the first time. These antisera were raised against the C-terminal octapeptide APSKDTVE, representing residues 121–128 of rCgA, Y⁰SGEATDGARPQALPEPMQESK, representing residues 124–144 of hCgA (Bachem; La Jolla, CA), and the dodecapeptide KVAHQLQALRRG, representing residues 437–448 of rCgA. hCgA_124–144 exhibits limited sequence homology with the corresponding rCgA sequence except in the sequence 132–143. Hence, the antiserum was expected to recognize the corresponding sequence in rCgA, which differs from the human in two positions only (F in rCgA, instead of L, in position 136 and K in rCgA, instead of M, in position 140). The peptides were synthesized by solid-phase methods using Fmoc strategy employing an automated peptide synthesizer (Fields and Noble 1990) and were purifed by reverse-phase HPLC. Their molecular mass numbers were verified by fast atom bombardment mass spectroscopy. They were coupled via glutaraldehyde to ovalbumin (rCgA_121–128), keyhole limpet hemocyanin (hCgA_124–144), and bovine serum albumin (Sigma; St Louis, MO) (rCgA_437–448). Rabbits received an SC priming injection of the conjugates in an emulsion of Freund's complete adjuvant (200 μg/ml). A second booster injection [100 μg/ml; Freund's complete:incomplete 1:9 (v/v)] was administered after 6 weeks, followed by additional monthly boosts. The rabbits were bled 2 weeks after each booster injection. Rabbits with code number R876 (antigen rCgA12_1–128), R36D8 (antigen hCgA_124–144), and R881 (antigen rCgA437–448) generated antisera suitable for immunohistochemistry (Table 1; Figure 1).

Table 1

Details of the antisera used a

Antisera raised to	Code no.	Working dilution	Reference/source
CgA_1–128	β-G	1:200	Curry et al. (1991)
bCgA_82–91	R880	1:800	Cunningham et al. (1996)
CgA_121–128	R876	1:800	See Materials and Methods
hCgA_124–144	R36D8	1:800	See Materials and Methods
CgA_152–240	H266	1:400	Curry et al. (1991)
CgA_241–306	H187	1:200	Curry et al. (1991)
CgA_291–319	A87A	1:200	Curry et al. (1991)
CgA_316–321	R633	1:800	Curry et al. (1991)
CgA_306–314	R9006	1:1000	Håkanson et al. (1995)
CgA_318–349	H98	1:200	Curry et al. (1991)
hCgA_343–356	R814	1:200	Barkatullah et al. (1997)
CgA_351–356	R635	1:200	Curry et al. (1991)
CgA_375–446	SP187	1:400	Curry et al. (1991)
CgA_437–448	R881	1:200	See Materials and Methods
HDC	M9501	1:500	Dartsch and Persson (1998)
Somatostatin	GP182	1:500	Maule et al. (1990)
Gastrin	GP360	1:1000	Eurodiagnostica
Serotonin	R448	1:500	Maule et al. (1990)

^aAntisera to CgA fragments were raised in rabbits. All other antisera were raised in guinea pigs except for the serotonin antiserum, R448, which was raised in a rabbit.

Markers for Identification of Individual Endocrine Cell Populations

An antiserum raised in a guinea pig against rat HDC (code no. M9501; Dartsch and Persson 1998; a kind gift from Dr L. Persson, Institute of Physiological Sciences, University of Lund, Sweden), was used as an ECL cell marker (Håkanson et al. 1986, 1994). Guinea pig antisera against rat somatostatin (GP182) (Maule et al. 1990) and rat gastrin (GP360) (Eurodiagnostica; Malmö, Sweden) were used to immunostain D and G cells, respectively. A rabbit antiserum against serotonin (R448) was used to immunostain EC cells. A-like cells were identified in an indirect manner, i.e., as round epithelial cells in the oxyntic mucosa that contained CgA-derived peptides but failed to react with antiserum to either HDC or somatostatin.

Animals

Male Sprague-Dawley rats weighing 200–250 g were sacrificed by exsanguination from the abdominal aorta under chloral hydrate anesthesia (300 mg/kg, IP). The stomach was opened along the greater curvature, washed in 0.9% saline, and placed, mucosal side up, on a cold glass surface. Tissue specimens from the acid-producing part of the stomach were taken from the ventral side close to the rumen. Specimens from the antrum were collected from the ventral side 1–2 mm above the pyloric sphincter.

Figure 1

Rat CgA (1–448) with pairs of basic amino acids (representing potential cleavage sites, arrows). Dotted lines indicate CgA sequence used for immunisation and unbroken horizontal lines indicate identified epitopes for the various antisera. The code number of each antiserum is shown above its respective antigen. Well-known CgA-derived peptides are shown for comparison (bottom line).

Immunocytochemistry

Tissue specimens were immersion-fixed in 4% paraformaldehyde in 0.1 M PBS, (pH 7.2) for 20 hr at 4C and cryoprotected in 5% (w/v) and 30% sucrose in 0.1 M phosphate buffer containing 0.01% NaN₃ before freezing in OCT embedding medium (Miles; Elkhart, IN). Frozen sections (6 μm) were cut using a Leitz cryostat and incubated (18 hr, 4C) with one of the primary antisera (Table 1). Immunoreaction was visualized using rhodamine-conjugated secondary swine anti-rabbit or fluorescein-conjugated rabbit anti-guinea pig IgG antibodies (1 hr, room temperature). Co-localization studies were performed as described previously (Curry et al. 1991). Briefly, sections were incubated with both primary guinea pig and primary rabbit antisera (Table 1) overnight. The sections were then washed in PBS (10 min) and incubated for 60 min with rhodamine-conjugated swine anti-rabbit antiserum, then with normal rabbit serum (0.1% in PBS, 30 min), followed by fluorescein-conjugated rabbit anti-guinea pig antiserum (60 min). Co-localization of antisera against CgA fragments and the rabbit antiserum against serotonin was examined by serial sectioning (4-μm-thick sections). Immunostaining was examined using a MRC 500 confocal scanning laser microscope (Bio-Rad Lasersharp; Richmond, CA) and an Olympus BH2 light microscope.

Control studies included liquid-phase preabsorption of each antiserum with its respective antigen (Table 1) and with the antigens used to raise the other antisera to CgA-derived peptides, and also the routine omission of primary antisera and substitution with non-immune rabbit, rat, and guinea pig serum. Truncated and C-terminally extended fragments of the respective antigens were synthesized (as above) and used to further characterize the epitope recognized by the various antisera (Table 2). These fragments included ELSEVLE, ELSEVLEN (of antigen YEDELSEVLE), SKDTVE (of antigen APSKDTVE), QELEK (of antigen Y⁰KGQELE), LGPPQGLFPG-amide, GPPQGLFPG, PPQGLFPG-amide, PQGLFPG-amide and GPPQGLFP (of antigen GPPQGLFPG-amide), WSKMDQLA, SKMDQLAKE, and ELTAEK (of antigens WSKMDQLAKELTAE (WE-14; Curry et al. 1992) and Y⁰KELTAE), and KVAHQLQALRRG and KVAHQLQALRR (of antigen Y⁰KVAHQLQALRRG). Fragments of antigen hCGA_124–144 were derived by proteolytic cleavage. Briefly, 500 μg of the synthetic peptide Y⁰SGEATDGARPQALPEPMQESK (hCgA_124–144, 500 μg ml^–1) was subjected to trypsinization, 0.2 mg ml^–1, 14 hr, 37C (Trypsin type XIII, 10,400 BAEE units; Sigma, St. Louis, MO). The digest was subjected to reverse-phase HPLC, employing a C₁₈ column (4.6 × 250 mm) (Supelcosil Technology; Bellefonte, CA) and mass spectrometry with a Voyager-DE Biosystem. Purities >95% were achieved for all fragments. The masses of the peptide fragments were compared with those predicted for cleavage at specific residues. The following five peptide fragments were isolated and identified: Y⁰SGEATDGAR, Y⁰SGEATDGARPQ, Y⁰SGEATDGARPQALPEPM, PQALPEPMQESK, and ALPEPMQESK. Three fragments, Y⁰SGEATDGARPQALPEPM, PQALPEPMQESK, and ALPEPMQESK, were quantified by radioimmunoassay (Hogg et al. 1998) (the yield was 3, 5 and 11 μg, respectively). The yield of Y⁰SGEATDGAR and Y⁰SGEATDGARPQ was approximated (5 and 28 μg, respectively) by comparing the heights of the HPLC peaks (detector absorbance at 214 nm) to that of the synthetic peptide Y⁰SGEATDGARPQALPEPMQESK.

Results

Specificity of Antisera

No immunostaining was observed after omission of the primary antiserum and the inclusion of nonimmune rabbit or guinea pig serum. Liquid-phase preabsorption of each antiserum with its respective antigen (1 μg/ml, except for R876 100 μg/ml) overnight at 4C abolished immunostaining (Table 2). Preabsorption with antigens used to raise other CgA antisera did not affect immunostaining. Antisera were further characterized after preabsorption with truncated and/or N-terminally or C-terminally extended peptides to determine epitope recognition sites (Table 2). Antiserum R880, raised against bCgA82–91 (YEDELSEVLE), was shown to detect the mid- or C-terminal part of the antigen in that both ELSEVLE and ELSEVLEN abolished immunostaining (10 μg/ml). Immunostaining of antiserum R876 raised against CgA_121–128 (APSKDTVE, C-terminal end of β-granin; Hutton et al. 1987) was abolished by SKDTVE (100 μg/ml), suggesting that the antibody recognizes the C-terminus of the antigen. Antiserum R36D8, raised against hCgA_124–144 (Y⁰SGEATDGARPQALPEPMQESK), appears to detect the the mid-portion (ALPEPM) of the antigen in that Y⁰SGEATDGARPQALPEPM, PQALPEPMQESK, and ALPEPMQESK blocked the staining (1 μg/ml), whereas neither Y⁰SGEATDGAR nor Y⁰SGEATDGARPQ did so (nor did the first mentioned three fragments after methionine oxidization). R633, raised against CgA316–321 (Y⁰KGQELE), is directed against the free C-terminal end (QELE) in that QELEK failed to block the immunoreaction. Antiserum R9006, raised against CgA_306–314 (GPPQGLFPG-amide, C-terminal end of pancreastatin; Tatemoto et al. 1986), is directed against the C-terminal octapeptide PPQGLFPG in that LGPPQGLFPG-amide, GPPQGLFPG-amide and PPQGLFPG-amide blocked the reaction, whereas GPPQGLFP and PQGLFPG-amide did not. The antiserum does not require the C-terminal amide because non-amidated GPPQGLFPG blocked the reaction. R814, raised against WE-14 (hCgA343–356, WSKMDQLAKELTAE), recognizes the N-terminal portion of the antigen because SKMDQLAKE, Y⁰KELTAE, and ELTAEK failed to block the immunostaining [whereas WSKMDQLA did (100 μg/ml)]. R635, raised against the C-terminal region of WE-14 (CgA_351–356, Y⁰KELTAE), is directed against the free C-terminal end of ELTAE in that ELTAEK failed to block the immunoreaction. Likewise, antiserum R881, raised against CgA_437–448 (YKVAHQLQALRRG), requires the C-terminal glycine residue for recognition because KVAHQLQALRRG (1 μg/ml), but not KVAHQLQALRR, blocked the reaction.

Table 2

Characterization of the epitopes recognized by the various CgA antisera employing truncated and N- and C-terminally extended CgA fragments a

CgA sequence	Antibody code no.	Antigen	Peptide derivative	Absorption, peptide conc. (μg.mh^–1)	Immunostaining after absorption
bCgA_82–91	R880	YEDELSEVLE	ELSEVLE	10	−
		ELSEVLEN	10	−
rCgA_121–128	R876	APSKDTVE	SKDTVE	100	−
hCgA_124–144	R36D8	Y⁰SGEATDGARPQALPEPM-QESKRPQALPEPMQESK	Y⁰SGEATDGARPQALPEPM b	1	−
			PQALPEPMQESK b	1	−
			ALPEPMQESK b	1	−
			Y⁰SGEATDGAR b	100	+
			Y⁰SGEATDGARPQ b	100	+
rCgA_316–321	R633	Y⁰KGQELE	QELEK	100	+
rCgA_306–314	R9006	GPPQGLFPG-amide	LGPPQGLFPG-amide	1	−
			PPQGLFPG-amide	1	−
			PQGLFPG-amide	100	+
			GPPQGLFPG	1	−
			GPPQGLFP	100	+
hCgA_343–356	R814	WSKMDQLAKELTAE	Y⁰WSKMDQLAKELTAE	1	−
			WSKMDQLAKELTAEY	1	−
			WSKMDQLA	100	−
			SKMDQLAKE	100	+
			Y⁰KELTAE	100	+
			ELTAEK	100	+
rCgA_351–356	R635	Y⁰KELTAE	Y⁰WSKMDQLAKELTAE	1	−
			WSKMDQLAKELTAEY	100	+
			Y⁰KELTAE	1	−
			ELTAEK	100	+
rCgA_437–448	R881	Y⁰KVAHQLQALRRG	KVAHQLQALRR	1	−
			KVAHQLQALRRG	100	+

^aThe epitopes was not determined for six of the antisera (β-G, H266, H187, A87A, H98, and SP187).

^bPeptide fragments generated after tryptic digestion of the synthetic peptide hCgA_124–144.

Distribution of CgA Immunoreactivity

The pattern of immunostaining observed after application of the various CgA antisera to the different endocrine cell populations of the oxyntic mucosa and antrum are summarized in Table 3. Seven of the 14 antisera immunostained all of the major endocrine cell populations (ECL cells, A-like cells, and D cells) in the oxyntic mucosa, and nine immunostained all endocrine cell populations in the antrum (EC cells, G cells, and D cells). The remaining antisera displayed a selective pattern of immunostaining. In the oxyntic mucosa, the ECL cells, A-like cells, and D cells were immunonegative after incubation with antiserum R635 (raised against CgA_351–356). However, a discrete population of cells in the oxyntic mucosa, possibly D₁/P cells, did exhibit moderate immunostaining. In the antrum, this antiserum immunostained G cells exclusively.

Table 3

Immunostaining of endocrine cells in rat stomach using antisera raised against fragments of the CgA sequence

		Oxyntic mucosa			Antral mucosa
Antiserum	Antigen	ECL cell	A-like cell	D cell	EC cell	G cell	D cell
β-G	CgA_1–128	+	+	+	+	+	+
R880	bCgA_82–91	−	+	+	+	+	+
R876	CgA_121–128	−	+	+	+	+	+
R36D8	hCgA_124–144	+	+	+	+	+	+
H266	CgA_152–240	+	+	+	+	+	+
H187	CgA_241–306	+	+	+	+	+	+
A87A	CgA_291–319	−	−	+	−	−	+
R633	CgA_316–321	−	−	+	−	+	+
R9006	CgA_306–314	+	+	+	+	+	+
H98	CgA_318–349	+	+	−	+	+	−
R814	hCgA_343–356	+	+	+	+	+	+
R635	CgA_351–356	−	−	−	−	+	−
SP187	CgA_375–446	+	+	+	+	+	+
R881	CgA_437–448	+	+	−	+	+	−

Oxyntic Mucosa

ECL Cells. Nine of the antisera immunostained the ECL cells (Table 3), but none of them immunostained the ECL cells exclusively (Figure 2).

A-like Cells. Antisera R880 and R876 (raised against bCgA_82–91 and CgA_121–128, respectively) failed to stain the ECL cells, whereas D cells and an additional sub-population of endocrine cells did display staining. Double immunostaining with the somatostatin antiserum revealed that most of these immunopositive cells were distinct from the D cells. The immunopositive cells were larger than D cells, exhibited a distinct spherical shape, and were found throughout the glands of the oxyntic mucosa (Figures 3a and 3b). Their number and morphological features are reminiscent of the A-like cells (Sundler and Håkanson 1991). Eleven of the fourteen antisera immunostained this distinct cell population.

D Cells. Eleven of the 14 antisera demonstrated the D cells. Seven of these also immunostained the ECL cells, and nine immunostained the A-like cells. Three of the 14 antisera, H98 (raised against CgA_318–349), R635 (raised against CgA_351–356), and R881 (raised against CgA_437–448), failed to stain the D cell population. Double immunostaining with the somatostatin antiserum revealed that both A87A and R633 (raised against CgA_291–319 and CgA_316–321, respectively) immunostained D cells exclusively (Figures 3c and 3d).

Figure 2

Immunostaining of oxyntic mucosa with antiserum M9501 (raised against HDC) (a) and antiserum H187 (raised against rCgA_241–306) (b). In the oxyntic mucosa, antiserum H187 immunostained numerous ECL cells, mainly located in the basal half of the crypts, and a few other endocrine cells (arrows) (D cells and A-like cells). Bar = 100 μm.

Figure 3

(a) Immunostaining of oxyntic mucosa with antiserum M9501 (raised against HDC) (green) and antiserum R880 (raised against bCgA_82–91) (red). R880 failed to stain the ECL cells, whereas another subpopulation of endocrine cells (possibly A-like cells) did display immunostaining. (b) Double immunostaining of oxyntic mucosa with antiserum R880 (red) and the somatostatin antiserum (green) revealed that R880 immunostained D cells weakly. However, most of the immunopositive cells were distinct from the D cells. These cells were larger than the D cells, appeared to have a spherical shape, and were uniformly distributed in the basal two thirds of the crypts. They were tentatively identified as A-like cells. (c) Double immunostaining of oxyntic mucosa with antiserum R633 (raised against CgA_316–321) and the somatostatin antiserum (d). R633 immunostained D cells exclusively. Bar = 100 μm.

Antral Mucosa

G Cells. Antiserum R635 (raised against CgA_351–356), which failed to stain the ECL cells, A-like, or D cells in the oxyntic mucosa, immunostained the G cells in the antrum. Antiserum A87A (raised against CgA_291–319) failed to stain the G cell population. All other CgA antisera immunostained the G cells.

EC Cells. None of the antisera immunostained EC cells exclusively. Most antisera immunostained the G, D, and EC cell populations.

D Cells. As in the oxyntic mucosa, A87A (raised against CGA_291–319) immunostained D cells exclusively. Antisera H98, R635, and R881 (raised against CgA_318–349, CgA_351–356 and CgA_437–448, respectively), failed to stain D cells in the antrum as well as in the oxyntic mucosa.

Discussion

Both light microscopic and electron microscopic techniques have been employed to localize CgA and CgA-derived peptides to gastric endocrine cells (Facer et al. 1985; Varndell et al. 1985; Håkanson et al. 1986; Rindi et al. 1986; Buffa et al. 1988; Grube et al. 1989; Lamberts et al. 1990; Curry et al. 1991; Portela-Gomes et al. 1997). Although CgA has been claimed to be a “marker” common to all neuroendocrine cells, endocrine cells of the gut are known to display heterogeneity of CgA immunoreactivity (Cetin et al. 1989). Variation in CgA immunoreactivity has been noted from one cell type to another and from one species to another. Clearly, the biological role of CgA is not uniform among these various cell populations. These variations may reflect the expression of different sets of processing enzymes. Alternatively, the enzymes might be the same but their rate of processing may differ greatly between the various cell types. CgA is subjected to post-translational modifications such as proteolysis and peptide amidation, glycosylation, and phosphorylation (Tatemoto et al. 1986; Winkler et al. 1986; Gorr and Cohn 1990; Barbosa et al. 1991). If the post-translational modifications are cell-specific, different epitopes will be expressed in different cell types, which may explain why certain antisera immunostain some endocrine cell types but not others. Moreover, the post-translational modifications may have a significant impact on the substrate recognition of the proteolytic enzymes and hence on which CgA-derived peptides are generated in the different cell populations. However, discrepancies among reports from different laboratories may also reflect different methodology (fixation and further processing of samples; Van Ewijk et al. 1984) and/or differences among the antisera applied rather than cell-specific differences in post-translational processing. To exclude methodological variables, all tissue samples in the present study were fixed and processed in a standardized way, thus facilitating a direct comparison of the various gastric endocrine cell types with one another. Fourteen site-specific CgA antisera were employed, together with cell-specific markers, to characterize the CgA immunoreactants in the different endocrine cell populations of the rat stomach. Previous investigations have reported differences in the expression of CgA-derived peptides in endocrine cells of the gastrointestinal tract. However, the various endocrine cell populations were not identified (Curry et al. 1991; Watkinson et al. 1991).

Oxyntic Mucosa

The ECL cells represent the predominant endocrine cell type of the oxyntic mucosa. In the rat stomach, histamine is found in both mast cells and ECL cells (Håkanson et al. 1986; Nissinen and Panula 1993), but only the ECL cells contain immunoreactive HDC (Kubota et al. 1984; Håkanson et al. 1986; Andersson et al. 1996). Nine of the 14 antisera to the different CgA domains displayed co-localization of CgA and CgA-derived peptides with HDC. However, each of these antisera immunostained additional cell populations in the oxyntic mucosa. Therefore, none of the antisera demonstrated the ECL cells exclusively.

The somatostatin-secreting D cells are spindle-shaped cells with long slender projections (Alumets et al. 1979; Kusumoto et al. 1979; Larsson et al. 1979). In the oxyntic mucosa they are uniformly distributed throughout the crypts and are of the closed type. Previous studies failed to detect CgA-derived peptides in gastric D cells (Varndell et al. 1985; Buffa et al. 1988; Wiedenmann and Huttner 1989; Portela-Gomes et al. 1997). This might reflect the fact that D cells produce only small amounts of CgA or that the specificity of the antisera used was such that they failed to detect the epitopes presented by the D cells. In this study, 11 of the antisera immunostained the D cells as well as other endocrine cell types. Two antisera, R633 (raised against CgA_316–321) and A87A (raised against CgA_291–319), immunostained the D cells exclusively. Antiserum R633 is specific to the free C-terminus of CgA_316–321, which is generated by cleavage of dibasic amino acid residues CgA_322–323. It therefore appears that cleavage at this site occurs in D cells but not in other endocrine cells in the oxyntic mucosa, suggesting that D cells process CgA differently from ECL cells and A-like cells.

To date, there is no recognized immunohistochemical marker for A-like cells, and their identification is based on indirect criteria. A-like cells are CgA-immunoreactive cells that do not display either HDC or somatostatin immunoreactivity and that exhibit a characteristic spherical shape and occur dispersed throughout the gastric glands in lower numbers than the ECL cells. Eleven of the CgA antisera immunostained this endocrine cell population. However, no single antiserum immunostained these cells exclusively. Two antisera, R880 (raised against CgA82–91) and R876 (raised against CgA_121–128), which were raised against different regions of the N-terminal CgA domain (CgA_1–128), referred to as β-granin (Hutton et al. 1987), immunostained D cells and the A-like cells. In contrast, another antiserum (β-G), raised against full-length β-granin, immunostained all endocrine cell populations in this location. Antiserum H98 and the C-terminal antiserum R881 immunostained ECL cells and the putative A-like cells but not the D cells. Again, this differential pattern of immunostaining supports the concept of cell-specific proteolysis of CgA. Antiserum R635, which is specific for the free C-terminus of WE-14 (CgA343–356; Curry et al. 1992), immunostained the G cells (see below) but failed to immunostain either ECL cells, A-like cells, or D cells in the oxyntic mucosa, suggesting that neither of these cell populations is able to produce WE-14.

Antrum

The predominant endocrine cell populations of the antral mucosa, the G cells, EC cells, and D cells, are localized in the basal part of the glands. All three endocrine cell populations are of the “open” type. Antiserum R635, specific for the C-terminal end of WE-14, immunostained G cells but not EC cells or D cells. This observation suggests that WE-14 is generated in G cells only.

Antral D cells were immunostained by 11 of the 14 antisera raised against CgA. One antiserum, A87A (raised against CgA29_1–319), immunostained the antral (and oxyntic) D cells but no other endocrine cells. R633 (raised against CgA_316–321) immunostained antral (and oxyntic) D cells and G cells. Antiserum R36D8 (raised against CgA_124–144) immunostained EC cells and D cells but not G cells. The fact that the D cells of both the oxyntic mucosa (closed type) and the antrum (open type) displayed identical patterns of CgA immunostaining is notable in view of the fact that they appear to differ from each other both morphologically and functionally (Alumets et al. 1979; Olesen et al. 1987).

Eleven of the 14 antisera immunostained the EC cells. The EC cell pattern of CgA immunostaining was identical to that of the A-like cells in the oxyntic mucosa, implying similarities in biosynthesis and processing of CgA.

Conclusions

In this study, 14 well-characterized CgA antisera were tested to generate information on the extent of cell-specific processing in the endocrine cells of the rat stomach. All major endocrine cell types were found to express CgA or CgA-derived peptides. In addition, the application of region-specific antisera raised against a range of CgA-derived peptides has revealed that each endocrine cell type expresses a unique mixture of CgA-derived peptides. This implies that each endocrine cell type has a characteristic spectrum of processing enzymes, e.g., prohormone convertase, carboxypeptidase, and amidating enzymes. It cannot be excluded, however, that additional differences with respect to glycosylation, phosphorylation, sulfation, or the rate and degree of processing will influence the ability of the processed products to react with the antisera. We propose that antisera that recognize specific domains of CgA be used as tools to identify individual endocrine cell populations. This may aid in the identification of cell populations for which no cell-specific markers exist or which, for some reason, are difficult to immunostain with a cell-specific marker (such as D cells and G cells in somatostatin or gastrin knockout animals). The functional significance of the present findings may perhaps be discussed in relation to the prohormones produced by the various endocrine cell types (such as preprogastrin in G cells or prosomatostatin in D cells); the CgA processing pattern may predict how the prohormones are being processed.

Footnotes

Acknowledgments

Supported by grants from the Swedish MRC (04X-1007), the Cancer Foundation (2542-B95–07XCC), the A. Påhlsson Foundation, the Medical Faculty, (University of Lund; Lund, Sweden), and Ipsen Pharmaceuticals, (London, UK).

References

Alumets

Ekelund

Munshid

Håkanson

Lorén

Sundler

(1979) Topography of somatostatin cells in the stomach of the rat: possible functional significance. Cell Tissue Res 202:177–188

Andersson

Cabero

Mattsson

Håkanson

(1996) Gastric acid secretion after depletion of enterochromaffin-like cell histamine. A study with α-fluoromethylhistidine in rats. Scand J Gastroenterol 31:24–30

Arden

Rutherford

Guest

Curry

Bailyes

Johnston

Hutton

(1994) The post-translational processing of chromogranin A in the pancreatic islet: involvement of the eukaryote subtilisin PC2. Biochem J 298:521–528

Barbosa

Gill

Takiyyuddin

O'Connor

(1991) Chromogranin A: posttranslational modifications in secretory granules. Endocrinology 128:174–190

Barkatullah

Curry

Johnston

Hutton

Buchanan

(1997) Ontogenetic expression of chromogranin A and its derived peptides, WE-14 and pancreastatin, in the rat neuroendocrine system. Histochem Cell Biol 107:251–258

Buffa

Maré

Gini

Salvadore

(1988) Chromogranins A and B and secretogranin II in hormonally identified endocrine cells of the gut and pancreas. Bas Appl Histochem 32:471–484

Capella

Finzi

Cornaggia

Usellini

Luinetti

Buffa

Solcia

(1991) Ultrastructural typing of gastric endocrine cells. In Håkanson

Sundler

The Stomach as an Endocrine Organ. Fernström Foundation Symposium no. 15. Amsterdam, Elsevier Science, 27–51

Cetin

Muller-Köppel

Aunis

Bader

M-F

Grube

(1989) Chromogranin A (CgA) in the gastro-entero-pancreatic (GEP) endocrine system. II. CgA in mammalian entero-endocrine cells. Histochemistry 92:265–275

Cunningham

Pogue

Curry

Johnston

Buchanan

(1996) PC-12 cells show immunoreactivity to a number of proteins and peptides including the vasostatins. Peptides 17:1297–1301

10.

Curry

Johnston

Hutton

Arden

Rutherford

Shaw

Buchanan

(1991) The tissue distribution of rat chromogranin A-derived peptides: evidence for differential tissue processing from sequence specific antisera. Histochemistry 96:531–538

11.

Curry

Shaw

Johnston

Thim

Buchanan

(1992) Isolation and primary structure of a novel chromogranin A-derived peptide, WE-14, from a human midgut carcinoid tumor. FEBS Lett 301:319–321

12.

Dartsch

Persson

(1998) Recombinant expression of rat histidine decarboxylase: generation of antibodies useful for Western blot analysis. Int J Biochem Cell Biol 30:773–782

13.

Dockray

Varro

Watkinson

Dimaline

(1991) Selective processing of peptides in gastric endocrine cells. In Håkanson

Sundler

The Stomach as an Endocrine Organ. Fernström Foundation Symposium no. 15. Amsterdam, Elsevier Science, 197–210

14.

Ekelund

Håkanson

Hedenbro

Rehfeld

Sundler

(1985) Endocrine cells and parietal cells in the stomach of the developing rat. Acta Physiol Scand 124:483–497

15.

Facer

Bishop

Lloyd

Wilson

Hennessy

Polak

(1985) Chromogranin: a newly recognised marker for endocrine cells of the human gastrointestinal tract. Gastroenterology 89:1366–1373

16.

Fields

Noble

(1990) Solid phase peptide synthesis utilising a fluorenylmethoxycarbonyl amino acids. Int J Pept Protein Res 35:161–214

17.

Fischer-Colbrie

Lassmann

Hagn

Winkler

(1985) Immunological studies on the distribution of chromogranin A and B in endocrine and nervous tissues. Neurscience 16:547–555

18.

Fischer-Colbrie

Hagn

Schober

(1987) Chromogranins A, B and C: widespread constituents of secretory vesicles. Ann NY Acad Sci 493:120–134

19.

Gorr

Cohn

(1990) Secretion of sulfated and nonsulfated forms of parathyroid chromogranin A (secretory protein-I). J Biol Chem 265:3012–3016

20.

Grube

Bargsten

Cetin

Yoshie

(1989) Chromogranins in mammalian GEP endocrine cells: their distribution and interrelations with co-stored amines and peptides. Arch Histol Cytol 52(suppl):91–98

21.

Håkanson

Böttcher

Ekblad

Panula

Simonson

Dohlsten

Hallberg

Sundler

(1986) Histamine in endocrine cells in the stomach: a survey of several species using a panel of histamine antibodies. Histochemistry 86:5–17

22.

Håkanson

Chen

Sundler

(1994) The ECL cells. In Johnson

The Physiology of the Gastrointestinal Tract. New York, Raven Press, 1171–1184

23.

Håkanson

Ding

Norlén

Chen

(1995) Circulating pancreastatin is a marker for the enterochromaffin-like cells of the rat stomach. Gastroenterology 108:1445–1456

24.

Helle

(1990) Chromogranins: universal proteins in secretory organelles from paramecium to man. Neurochem Int 17:165–175

25.

Hogg

Curry

Johnston

Buchanan

(1998) Quantification of total chromogranin A expression in human neuroendocrine tissues. Dig Dis Sci 43:1877

26.

Hutton

Hansen

Peshavaria

(1987) Proteolytic processing of chromogranin A in purified insulin granules. Formation of a 20 kDa N-terminal fragment (betagranin) by the concerted action of a Ca²⁺-dependent endopeptidase and carboxypeptidase. Biochem J 244:457–464

27.

Hutton

Nielsen

Kastern

(1988) The molecular cloning of the chromogranin A-like precursor of β-granin and pancreastatin from the endocrine pancreas. FEBS Lett 236:269–274

28.

Iacangelo

Okyama

Eiden

(1988) Primary structure of rat chromogranin A and distribution of its mRNA. FEBS Lett 227:115–121

29.

Kubota

Taguchi

Tohyama

Matsuura

Shiosaka

Ishihara

Watanabe

Shiotani

Wada

(1984) Electron microscopic identification of histidine decarboxylase-containing endocrine cells of the rat gastric mucosa. An immunohistochemical analysis. Gastroenterology 87:496–502

30.

Kusumoto

Iwanaga

Ito

Fujita

(1979) Juxtaposition of somatostatin cell and parietal cell in the dog stomach. Arch Histol Jpn 42:459–465

31.

Lamberts

Schmidt

Creutzfeldt

(1990) Light and electron microscopical immunocytochemical localization of pancreastatin-like immunoreactivity in porcine tissues. Histochemistry 93:369–380

32.

Larsson

Goltermann

De Magistris

Rehfeld

Schwartz

(1979) Somatostatin cell processes as pathways for paracrine secretion. Science 205:1393–1395

33.

Maule

Halton

Johnston

Shaw

Fairweather

(1990) A cytochemical study of the serotoninergic, cholinergic and peptidergic components of the reproductive system in the monogenean parasite, Diclidophora merlangi. Parasitol Res 76:409–419

34.

Nissinen

Panula

(1993) Histamine-storing cells in the oxyntic mucosa of the rat stomach: a transmission electron microscopic study employing fixation with carbodimide. J Histochem Cytochem 41:1405–1412

35.

O'Connor

(1983) Chromogranin: widespread immunoreactivity in polypeptide hormone producing tissues and in serum. Regul Pept 6:263–280

36.

O'Connor

Burton

Deftos

(1983) Chromogranin A: immunohistology reveals its universal occurrence in normal polypeptide hormone-producing endocrine glands. Life Sci 33:1657–1663

37.

O'Connor

Frigon

(1984) Chromogranin A, the major catecholamine storage vesicle soluble protein. J Biol Chem 259:3237–3247

38.

Olesen

Holst

Sottimano

Nielsen

(1987) Autonomic nervous control of fundic secretion of somatostatin and antral secretion of gastrin and somatostatin in pigs. Digestion 36:24–35

39.

Portela-Gomes

Stridsberg

Johansson

Grimelius

(1997) Complex co-localization of chromogranins and neurohormones in the human gastrointestinal tract. J Histochem Cytochem 45:815–822

40.

Rindi

Buffa

Sessa

Tortora

Solcia

(1986) Chromogranin A, B and C immunoreactivities of mammalian endocrine cells. Distribution, distinction from costored hormones/prohormones and relationship with the argyrophil component of secretory granules. Histochemistry 85:19–28

41.

Sandvik

Waldum

(1991) CCK-B (gastrin) receptor regulates gastric histamine release and acid secretion. Am J Physiol 260:G925–928

42.

Solcia

Capella

Vassallo

Buffa

(1975) Endocrine cells of the gastric mucosa. Int Rev Cytol 42:223–286

43.

Sundler

Håkanson

(1991) Gastric endocrine cell typing at the light microscopic level. In Håkanson

Sundler

The Stomach as an Endocrine Organ. Fernström Foundation Symposium no. 15. Amsterdam, Elsevier Science, 9–26

44.

Tatemoto

Efendic

Mutt

Makk

Feistner

Barchas

(1986) Pancreastatin, a novel pancreatic peptide that inhibits insulin secretion. Nature 324:476–478

45.

Van Ewijk

Van Soest

Verkerk

Jongkind

(1984) Loss of antibody binding to prefixed cells: fixation parameters for immunocytochemistry. Histochem J 16:179–193

46.

Varndell

Lloyd

Wilson

Polak

(1985) Ultrastructural localization of chromogranin: a potential marker for the electron microscopical recognition of endocrine cell secretory granules. Histochem J 17:981–992

47.

Watkinson

Jönsson

Davison

Young

Lee

Moore

Dockray

(1991) Heterogeneity of chromogranin A-derived peptides in bovine gut, pancreas and adrenal medulla. Biochem J 276:471–479

48.

Wiedenmann

Huttner

(1989) Synaptophysin and chromogranins/secretogranins. Widespread constituents of distinct types of neuroendocrine vesicles and new tools in tumor diagnosis. Virchows Arch [B] 58:95–121

49.

Winkler

Apps

Fischer-Colbrie

(1986) The molecular function of adrenal chromaffin granules: established facts and unresolved topics. Neuroscience 18:261–290

Cell-specific Processing of Chromogranin A in Endocrine Cells of the Rat Stomach

Abstract

Keywords

Materials and Methods

Antisera Production

Markers for Identification of Individual Endocrine Cell Populations

Animals

Immunocytochemistry

Results

Specificity of Antisera

Distribution of CgA Immunoreactivity

Oxyntic Mucosa

Antral Mucosa

Discussion

Oxyntic Mucosa

Antrum

Conclusions

Footnotes

Acknowledgments

References