0022-1554/92/$3.30 The Journal of Histochemistry and Cytochemistry Copyright 0 1992 by The Histochemical Society, Inc.
Vol. 40, No. 11, pp. 1619-1664, 1992 Printed in US.A.
Original Article
Light Microscopic Immunohistochemical Analysis of the Distribution of Group I1 Phospholipase A2 in Human Digestive Organs HIDE0 KIYOHARA, HIROSHI EGAMI, YUJI SHIBATA, KAZUYA MURATA, SHIGEKI OHSHIMA, and MICHIO OGAWA' Department of Surgety II, Kumamoto University Medical School, Kumamoto 860, Japan. Received for publication February 11, 1992 and in revised form June 18, 1992; accepted June 20, 1992 (2A2592).
The light microscopic and immunohistochemicaldistribution of human Group II phospholipase A2 (M-PLA2) in digestive organs of both human fetus and adult, with a new monoclonal antibody (MAb) against M-PLA2, was investigated semiquantitatively.The immunoreactivity was distributed similarly in the adult and fetal epithelium of the esophagus, duodenum, and small intestine, and in the acinar, islet, and duct cells of the pancreas. The epithelium of adult gallbladder was immunoreactive. Paneth cells, especially the secretory apparatus, were strongly immunoreactive. Hepatic Kiipffer cells and macrophagesof the adult spleen were also immunoreactive. These results suggest that these cells contain secretory-type Group I1 PLAz, which may be involved in host defensive mechanisms, such as phagocytosis in hu-
Introduction PhospholipaseA2 (PLA2) is a lipolytic enzyme which catalyzes the specific hydrolysis of a fatty acyl ester bond at the m - 2 position of 3-glycerophospholipids(1). Recent studies have demonstrated that cellular calcium-dependent PLA2 of mammalian origin can be classified into at least two types according to distinct characteristics of its primary structure (2): pancreatic-type PLA2 (Group I PLA2) and viperid/crotalid-type PLA2 [Group I1 PLA2 or membrane-associated PLAz (M-PLA2) (3,4)]. The latter has also been identified in intra-articular fluid ( 5 ) . Group I1 PLAz is hypothesized to be a regulatory factor in several metabolic pathways. Knowledge of the tissue and cellular distributions of this enzyme would assist in clarifying its biological function. However, the distribution of this enzyme in human organs has not been investigated. Recently, a murine monoclonal antibody (MAb)against human M-PLA2 was developed (6). In the present study, MAb against human M-PLA2 was used to hmunohistochemically analyze the dis-
Correspondence to: Michio Ogawa, Kumamoto Univ. Medical School, 1-1-1 Honjo, Kumamoto 860, Japan.
man digestive organs. In the adult colon, the immunoreactivity was observed only in the ascending colon and was not found in the transverse, descending, sigmoid, or rectal colon. The immunoreactivity was not found in fetal colon. Similarly, immunoreactivity was found in hepatocytes and Kiipffer cells of adult liver but not in fetal liver. By contrast, strong hunoreactivity was observed in the epithelium of the fetal stomach but not in adult stomach except in gastric neck cells. This suggests that the expression of M-PLA2 may be related to cell differentiation in particular organs. (JH i s t d e m Cytochem 40:1659-1&, 1992) KEY WORDS: Immunohistochemistry; M-PLA2;
Monoclonal anti-
body; Human; Digestive organs; Fetus; Adult.
tribution of M-PLA2 in human fetal and adult digestive organs including the esophagus, stomach, duodenum, small intestine, colon, rectum, liver, gallbladder, pancreas, and spleen.
Materials and Methods Tissue Samples. Normal adult tissues were obtained at autopsy from three patients (Case 1, 64 years, male, lung cancer and bronchitis; Case 2 . 6 9 years, female, bronchopneumonia; and Case 3, 75 years, male, laryngeal cancer and pneumonia) and surgical biopsy specimens were obtained from resections for neoplasms of adjacent organs. These biopsy specimens were obtained from one patient with esophageal cancer (65 years, male), seven with gastric cancer (age range 53-81 years, six males and one female), one with leiomyosarcoma of small intestine (39 years, female), nine with colon cancer (age range 56-73 years, four males and five females), and five with pancreatic cancer (age range 51-78 years, two males and three females). All tissues of the liver, gallbladder, and spleen used in this study were from autopsy specimens. The tissues studied were the esophagus, stomach, duodenum, jejunum, ileum, ascending colon, transverse colon, descending colon, sigmoid colon, rectum, liver, gallbladder, pancreas, and spleen. Adult digestive organs from at least three patients were studied. Fetal tissues were obtained at 23 and 25 weeks of gestation from necropsy specimens and consisted of the esophagus, stomach, small intestine, colon, liver, pancreas, and spleen. Each specimen was fixed in 20% buffered formaldehyde and embed-
1659
Downloaded from jhc.sagepub.com by guest on April 16, 2015
1660
KIYOHARA, EGMQ, SHIBATA, MURATA. OHSHIbfA, W A W A
Table 1. Dirttihion of M-Pur-imnrunoreache ceds in buman dgestive org&*b Pcmr
ortnn bPh?gUs Epithelium Gland Stomach Foveolar epithelium Gastric neck cell Gland Small intcniae Epithelium Pan& cell
2 3 d
25 w t c b
Mdc
2+
2+
n.a. n.0.
-
3+ n.0. 3+
-
3+ 3+
3+ 3+
n.c.
+
n.c.
n.c. n.c. n.c.
-
n.c.
n.c.
-
Figurarl-6 L i @ t m i u o a w p k i m m u n o h ~ ~ d a n W u m M M . P & MAb [dilutedconcentranon of 1:500 (488 r@nl)jAor human dlgesuw organ8 wHh the Min-blatln-perddtm complex mahod. No immurorr#divityWM fwnd in the mahodobgic and specifkity control reauio~.
-
2+
Figure 1. wultewphagw.Theepitheliumwa8 immunortwrctiva”ha!mom strongly posHive cdk, were found n the stratum corneum and 8Natum npim wm than in the stratum basale. 8ar 100 em.
n.a.
+
-
colon
Axcnding colon Tr~asvemcolon Descending colon Sigmoid colon Reed colon Liver Hepatocyte Ktipffer cell
n.c. n.c. n.c.
-
+
P9llCm.5
Acinar cell Iscler cell Ductnl cell Gallbladder Spleen LYmPhocytc MncrophPee
2+
3+
+ +
+ l2+ + /2+
n.a.
n.0.
3+
-
-
-
+
‘Fml tissues were obtained at 23 and 25 mb of gatnrion.
‘.
- no immunomcuve ccus WIC ohr~md;+ ,~ aCr ~ M 25% of ~ ~ I WIC I S immunomctive, 2 + .25-50% deck WCIC i m m u n o d v e ; 3 + more than 50%
.
ofccllr WICimmunodve; n.0.. tLara were not o b s e d : 0.a.. tisfua WICnot ani&ble: n.c.. riuua were not durified.
ded in @in. Serial 4-pm sectionswere obtained and one was stained with hematoxylin and eosin (HE) for routine histologic examination. The remaining scdons MIT immunohinodremicnllystained with anti-M-PLAz MAL. The C C i ”~ ~ n 0 ~ e n C t ifor ~ M-PLAz =KC chaterized with reference to the corrcsponding HE-stained slide.
I--
Anribody. Monoclonal antibody (MAL) was prepared by the method r e p m d by nhler and Milstein (7). Bdblc mice were immunized with M-PLAz purified from human spleen. MAb-producing hybridomas warp+ by fusioa ofspleen cek from immunized mice and the mouse mycloma cell line NS-1. The hybridomas were further propagated in mice as intraperitonealtumors. MAb ( 1 6 ) was purified from the ascites fluid of tumor-beoring mice by pt~teinA pffinityChromntogrpphy (6).
I ” & . The avidin-biorin-pcmxidase compla (ABC) method (8) was empbycd h a g the V d ABC Kit (Vector Labomtoria;Burling~me,CA). B~idy.Pnrpffinebeddedsections depnmff ~ z e in d xylene and hydrated in @ed solutionsof alcohol. Endogenour peroxidase acrivity was blocked by application of absolute methanol containing0.3% (w/v) hydrogen pcnxide for 30 min. T i e sections were then incubated in n o d hone serum (1:5%) for 30 min at mom temperanuc. E d spc~imCnwas h b a d with Optimpuy diluted pP~U-M-PLAZ MAL in a moist chamber =might at 4’C. A dilutionof MOO (4.86 ~ / d ‘RPS opdmal for in thi~ Study. Seleccion~wcw d d in 0.01 Mllin r PBS,pH 7.2. and biotinyhtcd hone anti-mow lgG (Vmor) wps ap-
-
plied at a dilution of 1:200 for 50 min at m m temperam. Freshly prepared ABC reagent was applied and incubated for 60min after three wrrha in PBS. The slides were r i d and the localization of M-PLAz was visualized by incubating the sections for 5 min in fmhly prepnrrd 0.05 Mlliter E%-HW. pH 7.6. containing 0.02% (wlv) 3,3’diaminobenzidinetcaahydrochloride (NPCPIPi k q u e ; Kyoto,Japan) and 0.03% (wlv) hydrogen pemxidc. The sections were countemninedwith Maycr‘s hematoxylin. The methodologic and specificity control reactions were p+ as follows: (a) sections were pnmsed without primary antibody. and (b) nonspecific m o w IgG replaced m o w anti-human M-PLAz MAb.
I-- Lnmanorr?cri*iq. M-PLAz immunodvity was asscscd remiquMtimtively pccording to the pemnugc d M-PLAz-immunorencdR ceb.Thegroupsiadudd grcptcr than 50% (3+. SnOnglyposk); 1O-25% (2 +, moderately positive); less than 25% ( + , d y positive); and 0% ( - , negative).
of Cek b&hi06 Fhttan. The light ~~UUSCOP~C Smimunoreactive for M-PLAz were ckrsificd as cytoplasmic pular, Golgi, or luminal surface patterns. A cytoplasmic granular mining panun wps defined as the presence of f i e granular m a d d u p d &rou&out the cytoplasm. When the stained coarse granules were confiicd to the supranuclear region ofthe cell, it was described as a Golgi pattem (9). Smining of the luminal surface of the cek was termed a luminal nubrae paturn.
Results
--
D&dution of M-PLA2 in Fetal a d Ad& Tusnes The distribution of the M-PLA2-immunodvc cells io human in able 1. IOthe ddt, digadVCOW k
tivcccllswctcvisupLitcdinthe~o~o~~ed~ )immunoreactivity was found in the small intestine and poncrrru. sevcrpl OW in imm~norrp~tiviy ddt and frtus. Immunodvc cells wcrc not o b s e d in fetal colon. liwr,
Downloaded from jhc.sagepub.com by guest on April 16, 2015
M-PLA2 DISTRIBUTION IN HUMAN DIGESTIVE ORGANS
1661
-~
Figure 2. Adult stomach. In the normal epithelium and glands only gastric neck cells were immunoreactive in a Golgi pattern. Bar = 100 pm. Figure 3. Adult duodenum. (A) The epithelium was immunoreactive in a Golgi pattern. (B) The brush borders were strongly immunoreactive in a luminal surface pattern. (C) Paneth cells were particularly immunoreactive, which corresponded to the secretory apparatus (arrows). Crypt cells were also immunoreactive in a Golgi pattern. Bars = 100 pm. Figure 4. (A) Adult ascending colon. The epitheliumwas immunoreactive in a Golgi pattern. (E) Adult descending colon. No immunoreactive cells were observed. Bars = 100 um.
Downloaded from jhc.sagepub.com by guest on April 16, 2015
1662
KIYOHARA, EGAMI, SHIBATA, MURATA, OHSHIMA, OGAWA
..
. -
.
-. -
.
._ .
Figure 5. Adult pancreas. (A) Almost all acinar cells were immunoreactive in a Golgi pattern. Some islet cells (B) and some ductal cells (C) were also immunoreactive in a Golgi pattern. Bars = 100 pm. Figure 6. Adult liver. (A) The immunoreactivity was found in limited numbers of hepatocytes in a cytoplasmic granular pattern, with a zonular distribution in the lobule. (B) A Kupffer cell (arrow) was also immunoreactive in a cytoplasmic granular pattern. Bars: A = 200 pm; B = 50 pm.
or spleen. By contrast, strong immunoreactivity was observed in fetal but not adult gastric mucosa. There was no significant difference of the immunoreactivity between autopsy and surgical biopsy specimens and between sexes
examined in the present study. No immunoreactivity was found in the methodologic and specificity control reactions. Esophagus. The stratified epithelium of the esophagus was im-
Downloaded from jhc.sagepub.com by guest on April 16, 2015
M-PLAz DISTRIBUTION IN HUMAN DIGESTIVE ORGANS
munoreactive for M-PLA2. Strong immunoreactivity was particularly evident in the stratum corneum and stratum spinosum. In the stratum basale, few immunoreactive cells were found. There were no immunoreactive cells in the esophageal glands (Figure 1). The intensity and pattern of staining in the fetus was similar to that seen in the adult. Stomach. In the adult stomach there were no immunoreactive cells in either the normal epithelium or glands other than in the gastric neck cells (Figure 2). However, strong immunoreactivity was found in the epithelium and fundic glands of the fetus. Duodenum. The epithelium of the duodenum (Figure 3A) and Brunner’s glands were immunoreactive in a Golgi pattem, and brush borders (Figure 3B) showed strong immunoreactivity in a luminal surface pattern. Paneth cells (Figure 3C) were particularly immunoreactive, which corresponded to the supranuclearsecretory apparatus, and crypt cells were also immunoreactive in a Golgi pattern (Figure 3C). Small Intestine. In the adult, the epithelium of the jejunum and ileum was immunoreactive in a Golgi pattern. In the fetus, the epithelium of small intestine was also immunoreactive in a Golgi pattern. Brush borders and Paneth cells were strongly immunoreactive. Colon and Rectum. The epithelium of the adult ascending colon was immunoreactive in a Golgi pattern (Figure 4A). By contrast, immunoreactivity was not found in the transverse colon, descending colon (Figure 4B),sigmoid colon, or rectum of the adult. There were no immunoreactive cells in the colon of the fetus. Pancreas. Almost all acinar cells (Figure 5A) and some islet (Figure 5B)and duct cells (Figure 5C) were immunoreactivein a Golgi pattern in the adult. Immunoreactivity was also observed in the fetus, but the staining intensity was weaker than that seen in the adult. Liver. Immunoreactivitywas found in limited numbers of adult hepatocytes in a cytoplasmic granular pattern (Figure 6A), and a zonular distribution of immunoreactive hepatocytes in the lobule was observed. Kiipffer cells were also immunoreactive in a cytoplasmic granular pattern (Figure 6B). No immunoreactivecells were observed in fetal liver. Spleen. Adult macrophages were immunoreactive for M-PLA2 in a cytoplasmic granular pattern, whereas lymphocytes were devoid of immunoreactivity. In fetal spleen, no immunoreactivity was observed. Gallbladder. In the adult gallbladder, epithelium was immunoreactive in a Golgi pattern. Fetal gallbladder was not available for study.
1663
plete primary structure of human membrane-associated PLA2 (Group I1 PLA2) purified from a human spleen membrane fraction has been determined. The primary structure of secretory type human Group I1 PLAz has been reported and was confirmed to be identical to the primary structure of membrane-associated Group I1 PLAz (12). In the present study, M-PLA2 was observed to be widely distributed throughout various digestive organs and cell types in humans. In the adult, immunoreactivity for M-PLA2 was demonstrated in the epithelium of the esophagus, duodenum, small intestine, ascending colon and gall bladder, gastric neck cells, acinar, islet, and duct cells of pancreas, hepatocytes and Kiipffer cells of liver, and macrophages of spleen. The secretory apparatus of Paneth cells in the duodenum and small intestine showed strong immunoreactivityfor M-PLA2. Paneth cells are recognized to regulate intestinal flora through phagocytosis (13,14) and are involved in the secretion of lysozyme (15). The immunoreactivity found in the secretory apparatus of Paneth cells suggests that the M-PLA2 recognized by this MAb is secretory-type Group I1 PLA2. The expression of M-PLA2 in Paneth cells, Kiipffer cells, and macrophages suggests that human M-PLA2 plays an essential role in extramucosal immune defense mechanisms. Some differences in the presence of this enzyme were found between the fetus and the adult. M-PLA2 immunoreactivity was observed in the ascending colon of the adult but was not found in the fetal colon. Immunoreactivitywas also found in adult hepatocytes, Kiipffer cells, and spleen macrophages but not in corresponding fetal tissues. By contrast, strong immunoreactivity was observed in the epithelium of the fetal but not the adult stomach. In the adult stomach, M-PLA2 was retained only in gastric neck cells, which are known to be the proliferating cells of the gastric mucosa. In a separateinvestigation, we observed strongimmunoreactivity for M-PLA2 in the regenerating cell zone that develops subsequent to mucosal damage, and in intestinal metaplasia of crypt cells in the stomach. Furthermore, abundant expression of M-PLA2 was observed in poorly differentiated adenocarcinoma tissues, particularly in invasive gastric cancer (data not shown). These observations suggest that the expression of M-PLA2 is related to cell growth andlor differentiation. The present study revealed that M-PLA2 immunoreactivity is distributed throughout digestive organs. The distribution and intracellular localization of M-PLA2 in these organs suggests that M-PLA2 may have functions other than regulation of intracellular metabolism.
Acknowledgment We than&Dr U e y aforproviding adult autopsy andfetal necropsy specimens for this study.
Literature Cited
Discussion Membrane-associated-typeGroup I1 PLA2 is hypothesized to regulate cellular phospholipid metabolism and release of prostaglandin (PG) precursors (10). Secretory type Group I1 PLA2 is also secreted into intra-articular fluid ( 5 ) , and a PLA2 in rheumatoid arthritis synovial fluid has been partially sequenced (11). However, the biological function of this form is not clear. Recently, the com-
1. Vadas P, Pruzanski W. Biology of disease: role of secretory phospholi-
pase Az in the pathobiology of disease. Lab Invest 1986;55:391 2. Heinrikson RI., Krueger ET, Keim PS. Amino acid sequence of phospholipase A2-a from the venom of Crotalus adamanteus: a new classification of phospholipase A2 based upon structural determinants. J Biol Chem 1977;212:4913 3. Tojo H, Ono T, Kuramitsu S, Kagamiyama H, Okamoto M. A phos-
Downloaded from jhc.sagepub.com by guest on April 16, 2015
KIYOHARA, EGAMI, SHIBATA, MURATA, OHSHIMA, OGAWA
1664
pholipase A2 in the supernatant fraction of rat spleen: its similarity to rat pancreatic phospholipase A2. J Biol Chem 1988;263:5724 4. Ono T, Tojo H, Kuramitsu S, Kagamiyama H, Okamoto M. Purification and characterizationof a membrane-associated phospholipase A2 from rat spleen: its comparison with a cytosolic phospholipase A2 S-1. J Biol Chem 1988;263:5732 Purzanski W, Vadas P, Stefanski E, Urowitz MB. PhospholipaseA2 activity in sera and synovial fluids in rheumatoid arthritis and osteoarthritis: its possible role as a proinflammatory enzyme. J Rheumatol 1985;12:211 Matsuda Y, Ogawa M, Sakamoto K, Yamashita S, Kanda A. Kohno M, Yoshida N, Nishijima J, Murata A, Mori T. Development of a radioimmunoassay for human group I1 phospholipase A2 and demonstration of postoperative elevation. Enzyme, in press Kohler G, Milstein C. Continuous cultures of fused cells secreting antibody of predefined specificity. Nature 1975;256:495 Hsu SM, &ne L, Fanger H. The use of avidin-biotin-peroxidase complex (ABC) in immunoperoxidase techniques: a comparison between ABC and unlabeled antibody (PAP)procedures.J Histochem Cytochem 1982;29:577
9 Wiyama Y,Egami H, Pout PM. Expression of human tumor-associated antigens in pancreatic cancer induced in Syrian hamsters. Am J Pathol 1990;136:707 10 Van den Bosch H. Intracellular phospholipase A. Biochim Biophys Acta 1980;604:191 11. Vadas P, Stefanski E, Purzanski W. Characterizationof extracellularphospholipase A2 in rheumatoid synovial fluid. Life Sci 1985;36:579 12 Kanda A, Ono T, Yoshida N, Tojo H, Okamota M. The primary structure of a membrane-associatedphospholipase A2 from human spleen. Biochem Biophys Res Commun 1989;163:42 13 Erlandsen SL, Chase DG. Paneth cell function: phagocytosis and intracellular digestion of intestinal microorganisms. I. Hexamita murk J Ultrastruct Res 1972;41:296 14 Erlandsen SL, Chase DG. Paneth cell function: phagocytosis and intracellular digestion of intestinal microorganisms. 11. Spiral microorganism. J Ultrastruct Res 1972;41:319 15 Chipman DM, Sharon N. Mechanism of lysozyme action. Science 1966;19:540
Downloaded from jhc.sagepub.com by guest on April 16, 2015
Vol. 40, No. 11, pp. 1619-1664, 1992 Printed in US.A.
Original Article
Light Microscopic Immunohistochemical Analysis of the Distribution of Group I1 Phospholipase A2 in Human Digestive Organs HIDE0 KIYOHARA, HIROSHI EGAMI, YUJI SHIBATA, KAZUYA MURATA, SHIGEKI OHSHIMA, and MICHIO OGAWA' Department of Surgety II, Kumamoto University Medical School, Kumamoto 860, Japan. Received for publication February 11, 1992 and in revised form June 18, 1992; accepted June 20, 1992 (2A2592).
The light microscopic and immunohistochemicaldistribution of human Group II phospholipase A2 (M-PLA2) in digestive organs of both human fetus and adult, with a new monoclonal antibody (MAb) against M-PLA2, was investigated semiquantitatively.The immunoreactivity was distributed similarly in the adult and fetal epithelium of the esophagus, duodenum, and small intestine, and in the acinar, islet, and duct cells of the pancreas. The epithelium of adult gallbladder was immunoreactive. Paneth cells, especially the secretory apparatus, were strongly immunoreactive. Hepatic Kiipffer cells and macrophagesof the adult spleen were also immunoreactive. These results suggest that these cells contain secretory-type Group I1 PLAz, which may be involved in host defensive mechanisms, such as phagocytosis in hu-
Introduction PhospholipaseA2 (PLA2) is a lipolytic enzyme which catalyzes the specific hydrolysis of a fatty acyl ester bond at the m - 2 position of 3-glycerophospholipids(1). Recent studies have demonstrated that cellular calcium-dependent PLA2 of mammalian origin can be classified into at least two types according to distinct characteristics of its primary structure (2): pancreatic-type PLA2 (Group I PLA2) and viperid/crotalid-type PLA2 [Group I1 PLA2 or membrane-associated PLAz (M-PLA2) (3,4)]. The latter has also been identified in intra-articular fluid ( 5 ) . Group I1 PLAz is hypothesized to be a regulatory factor in several metabolic pathways. Knowledge of the tissue and cellular distributions of this enzyme would assist in clarifying its biological function. However, the distribution of this enzyme in human organs has not been investigated. Recently, a murine monoclonal antibody (MAb)against human M-PLA2 was developed (6). In the present study, MAb against human M-PLA2 was used to hmunohistochemically analyze the dis-
Correspondence to: Michio Ogawa, Kumamoto Univ. Medical School, 1-1-1 Honjo, Kumamoto 860, Japan.
man digestive organs. In the adult colon, the immunoreactivity was observed only in the ascending colon and was not found in the transverse, descending, sigmoid, or rectal colon. The immunoreactivity was not found in fetal colon. Similarly, immunoreactivity was found in hepatocytes and Kiipffer cells of adult liver but not in fetal liver. By contrast, strong hunoreactivity was observed in the epithelium of the fetal stomach but not in adult stomach except in gastric neck cells. This suggests that the expression of M-PLA2 may be related to cell differentiation in particular organs. (JH i s t d e m Cytochem 40:1659-1&, 1992) KEY WORDS: Immunohistochemistry; M-PLA2;
Monoclonal anti-
body; Human; Digestive organs; Fetus; Adult.
tribution of M-PLA2 in human fetal and adult digestive organs including the esophagus, stomach, duodenum, small intestine, colon, rectum, liver, gallbladder, pancreas, and spleen.
Materials and Methods Tissue Samples. Normal adult tissues were obtained at autopsy from three patients (Case 1, 64 years, male, lung cancer and bronchitis; Case 2 . 6 9 years, female, bronchopneumonia; and Case 3, 75 years, male, laryngeal cancer and pneumonia) and surgical biopsy specimens were obtained from resections for neoplasms of adjacent organs. These biopsy specimens were obtained from one patient with esophageal cancer (65 years, male), seven with gastric cancer (age range 53-81 years, six males and one female), one with leiomyosarcoma of small intestine (39 years, female), nine with colon cancer (age range 56-73 years, four males and five females), and five with pancreatic cancer (age range 51-78 years, two males and three females). All tissues of the liver, gallbladder, and spleen used in this study were from autopsy specimens. The tissues studied were the esophagus, stomach, duodenum, jejunum, ileum, ascending colon, transverse colon, descending colon, sigmoid colon, rectum, liver, gallbladder, pancreas, and spleen. Adult digestive organs from at least three patients were studied. Fetal tissues were obtained at 23 and 25 weeks of gestation from necropsy specimens and consisted of the esophagus, stomach, small intestine, colon, liver, pancreas, and spleen. Each specimen was fixed in 20% buffered formaldehyde and embed-
1659
Downloaded from jhc.sagepub.com by guest on April 16, 2015
1660
KIYOHARA, EGMQ, SHIBATA, MURATA. OHSHIbfA, W A W A
Table 1. Dirttihion of M-Pur-imnrunoreache ceds in buman dgestive org&*b Pcmr
ortnn bPh?gUs Epithelium Gland Stomach Foveolar epithelium Gastric neck cell Gland Small intcniae Epithelium Pan& cell
2 3 d
25 w t c b
Mdc
2+
2+
n.a. n.0.
-
3+ n.0. 3+
-
3+ 3+
3+ 3+
n.c.
+
n.c.
n.c. n.c. n.c.
-
n.c.
n.c.
-
Figurarl-6 L i @ t m i u o a w p k i m m u n o h ~ ~ d a n W u m M M . P & MAb [dilutedconcentranon of 1:500 (488 r@nl)jAor human dlgesuw organ8 wHh the Min-blatln-perddtm complex mahod. No immurorr#divityWM fwnd in the mahodobgic and specifkity control reauio~.
-
2+
Figure 1. wultewphagw.Theepitheliumwa8 immunortwrctiva”ha!mom strongly posHive cdk, were found n the stratum corneum and 8Natum npim wm than in the stratum basale. 8ar 100 em.
n.a.
+
-
colon
Axcnding colon Tr~asvemcolon Descending colon Sigmoid colon Reed colon Liver Hepatocyte Ktipffer cell
n.c. n.c. n.c.
-
+
P9llCm.5
Acinar cell Iscler cell Ductnl cell Gallbladder Spleen LYmPhocytc MncrophPee
2+
3+
+ +
+ l2+ + /2+
n.a.
n.0.
3+
-
-
-
+
‘Fml tissues were obtained at 23 and 25 mb of gatnrion.
‘.
- no immunomcuve ccus WIC ohr~md;+ ,~ aCr ~ M 25% of ~ ~ I WIC I S immunomctive, 2 + .25-50% deck WCIC i m m u n o d v e ; 3 + more than 50%
.
ofccllr WICimmunodve; n.0.. tLara were not o b s e d : 0.a.. tisfua WICnot ani&ble: n.c.. riuua were not durified.
ded in @in. Serial 4-pm sectionswere obtained and one was stained with hematoxylin and eosin (HE) for routine histologic examination. The remaining scdons MIT immunohinodremicnllystained with anti-M-PLAz MAL. The C C i ”~ ~ n 0 ~ e n C t ifor ~ M-PLAz =KC chaterized with reference to the corrcsponding HE-stained slide.
I--
Anribody. Monoclonal antibody (MAL) was prepared by the method r e p m d by nhler and Milstein (7). Bdblc mice were immunized with M-PLAz purified from human spleen. MAb-producing hybridomas warp+ by fusioa ofspleen cek from immunized mice and the mouse mycloma cell line NS-1. The hybridomas were further propagated in mice as intraperitonealtumors. MAb ( 1 6 ) was purified from the ascites fluid of tumor-beoring mice by pt~teinA pffinityChromntogrpphy (6).
I ” & . The avidin-biorin-pcmxidase compla (ABC) method (8) was empbycd h a g the V d ABC Kit (Vector Labomtoria;Burling~me,CA). B~idy.Pnrpffinebeddedsections depnmff ~ z e in d xylene and hydrated in @ed solutionsof alcohol. Endogenour peroxidase acrivity was blocked by application of absolute methanol containing0.3% (w/v) hydrogen pcnxide for 30 min. T i e sections were then incubated in n o d hone serum (1:5%) for 30 min at mom temperanuc. E d spc~imCnwas h b a d with Optimpuy diluted pP~U-M-PLAZ MAL in a moist chamber =might at 4’C. A dilutionof MOO (4.86 ~ / d ‘RPS opdmal for in thi~ Study. Seleccion~wcw d d in 0.01 Mllin r PBS,pH 7.2. and biotinyhtcd hone anti-mow lgG (Vmor) wps ap-
-
plied at a dilution of 1:200 for 50 min at m m temperam. Freshly prepared ABC reagent was applied and incubated for 60min after three wrrha in PBS. The slides were r i d and the localization of M-PLAz was visualized by incubating the sections for 5 min in fmhly prepnrrd 0.05 Mlliter E%-HW. pH 7.6. containing 0.02% (wlv) 3,3’diaminobenzidinetcaahydrochloride (NPCPIPi k q u e ; Kyoto,Japan) and 0.03% (wlv) hydrogen pemxidc. The sections were countemninedwith Maycr‘s hematoxylin. The methodologic and specificity control reactions were p+ as follows: (a) sections were pnmsed without primary antibody. and (b) nonspecific m o w IgG replaced m o w anti-human M-PLAz MAb.
I-- Lnmanorr?cri*iq. M-PLAz immunodvity was asscscd remiquMtimtively pccording to the pemnugc d M-PLAz-immunorencdR ceb.Thegroupsiadudd grcptcr than 50% (3+. SnOnglyposk); 1O-25% (2 +, moderately positive); less than 25% ( + , d y positive); and 0% ( - , negative).
of Cek b&hi06 Fhttan. The light ~~UUSCOP~C Smimunoreactive for M-PLAz were ckrsificd as cytoplasmic pular, Golgi, or luminal surface patterns. A cytoplasmic granular mining panun wps defined as the presence of f i e granular m a d d u p d &rou&out the cytoplasm. When the stained coarse granules were confiicd to the supranuclear region ofthe cell, it was described as a Golgi pattem (9). Smining of the luminal surface of the cek was termed a luminal nubrae paturn.
Results
--
D&dution of M-PLA2 in Fetal a d Ad& Tusnes The distribution of the M-PLA2-immunodvc cells io human in able 1. IOthe ddt, digadVCOW k
tivcccllswctcvisupLitcdinthe~o~o~~ed~ )immunoreactivity was found in the small intestine and poncrrru. sevcrpl OW in imm~norrp~tiviy ddt and frtus. Immunodvc cells wcrc not o b s e d in fetal colon. liwr,
Downloaded from jhc.sagepub.com by guest on April 16, 2015
M-PLA2 DISTRIBUTION IN HUMAN DIGESTIVE ORGANS
1661
-~
Figure 2. Adult stomach. In the normal epithelium and glands only gastric neck cells were immunoreactive in a Golgi pattern. Bar = 100 pm. Figure 3. Adult duodenum. (A) The epithelium was immunoreactive in a Golgi pattern. (B) The brush borders were strongly immunoreactive in a luminal surface pattern. (C) Paneth cells were particularly immunoreactive, which corresponded to the secretory apparatus (arrows). Crypt cells were also immunoreactive in a Golgi pattern. Bars = 100 pm. Figure 4. (A) Adult ascending colon. The epitheliumwas immunoreactive in a Golgi pattern. (E) Adult descending colon. No immunoreactive cells were observed. Bars = 100 um.
Downloaded from jhc.sagepub.com by guest on April 16, 2015
1662
KIYOHARA, EGAMI, SHIBATA, MURATA, OHSHIMA, OGAWA
..
. -
.
-. -
.
._ .
Figure 5. Adult pancreas. (A) Almost all acinar cells were immunoreactive in a Golgi pattern. Some islet cells (B) and some ductal cells (C) were also immunoreactive in a Golgi pattern. Bars = 100 pm. Figure 6. Adult liver. (A) The immunoreactivity was found in limited numbers of hepatocytes in a cytoplasmic granular pattern, with a zonular distribution in the lobule. (B) A Kupffer cell (arrow) was also immunoreactive in a cytoplasmic granular pattern. Bars: A = 200 pm; B = 50 pm.
or spleen. By contrast, strong immunoreactivity was observed in fetal but not adult gastric mucosa. There was no significant difference of the immunoreactivity between autopsy and surgical biopsy specimens and between sexes
examined in the present study. No immunoreactivity was found in the methodologic and specificity control reactions. Esophagus. The stratified epithelium of the esophagus was im-
Downloaded from jhc.sagepub.com by guest on April 16, 2015
M-PLAz DISTRIBUTION IN HUMAN DIGESTIVE ORGANS
munoreactive for M-PLA2. Strong immunoreactivity was particularly evident in the stratum corneum and stratum spinosum. In the stratum basale, few immunoreactive cells were found. There were no immunoreactive cells in the esophageal glands (Figure 1). The intensity and pattern of staining in the fetus was similar to that seen in the adult. Stomach. In the adult stomach there were no immunoreactive cells in either the normal epithelium or glands other than in the gastric neck cells (Figure 2). However, strong immunoreactivity was found in the epithelium and fundic glands of the fetus. Duodenum. The epithelium of the duodenum (Figure 3A) and Brunner’s glands were immunoreactive in a Golgi pattem, and brush borders (Figure 3B) showed strong immunoreactivity in a luminal surface pattern. Paneth cells (Figure 3C) were particularly immunoreactive, which corresponded to the supranuclearsecretory apparatus, and crypt cells were also immunoreactive in a Golgi pattern (Figure 3C). Small Intestine. In the adult, the epithelium of the jejunum and ileum was immunoreactive in a Golgi pattern. In the fetus, the epithelium of small intestine was also immunoreactive in a Golgi pattern. Brush borders and Paneth cells were strongly immunoreactive. Colon and Rectum. The epithelium of the adult ascending colon was immunoreactive in a Golgi pattern (Figure 4A). By contrast, immunoreactivity was not found in the transverse colon, descending colon (Figure 4B),sigmoid colon, or rectum of the adult. There were no immunoreactive cells in the colon of the fetus. Pancreas. Almost all acinar cells (Figure 5A) and some islet (Figure 5B)and duct cells (Figure 5C) were immunoreactivein a Golgi pattern in the adult. Immunoreactivity was also observed in the fetus, but the staining intensity was weaker than that seen in the adult. Liver. Immunoreactivitywas found in limited numbers of adult hepatocytes in a cytoplasmic granular pattern (Figure 6A), and a zonular distribution of immunoreactive hepatocytes in the lobule was observed. Kiipffer cells were also immunoreactive in a cytoplasmic granular pattern (Figure 6B). No immunoreactivecells were observed in fetal liver. Spleen. Adult macrophages were immunoreactive for M-PLA2 in a cytoplasmic granular pattern, whereas lymphocytes were devoid of immunoreactivity. In fetal spleen, no immunoreactivity was observed. Gallbladder. In the adult gallbladder, epithelium was immunoreactive in a Golgi pattern. Fetal gallbladder was not available for study.
1663
plete primary structure of human membrane-associated PLA2 (Group I1 PLA2) purified from a human spleen membrane fraction has been determined. The primary structure of secretory type human Group I1 PLAz has been reported and was confirmed to be identical to the primary structure of membrane-associated Group I1 PLAz (12). In the present study, M-PLA2 was observed to be widely distributed throughout various digestive organs and cell types in humans. In the adult, immunoreactivity for M-PLA2 was demonstrated in the epithelium of the esophagus, duodenum, small intestine, ascending colon and gall bladder, gastric neck cells, acinar, islet, and duct cells of pancreas, hepatocytes and Kiipffer cells of liver, and macrophages of spleen. The secretory apparatus of Paneth cells in the duodenum and small intestine showed strong immunoreactivityfor M-PLA2. Paneth cells are recognized to regulate intestinal flora through phagocytosis (13,14) and are involved in the secretion of lysozyme (15). The immunoreactivity found in the secretory apparatus of Paneth cells suggests that the M-PLA2 recognized by this MAb is secretory-type Group I1 PLA2. The expression of M-PLA2 in Paneth cells, Kiipffer cells, and macrophages suggests that human M-PLA2 plays an essential role in extramucosal immune defense mechanisms. Some differences in the presence of this enzyme were found between the fetus and the adult. M-PLA2 immunoreactivity was observed in the ascending colon of the adult but was not found in the fetal colon. Immunoreactivitywas also found in adult hepatocytes, Kiipffer cells, and spleen macrophages but not in corresponding fetal tissues. By contrast, strong immunoreactivity was observed in the epithelium of the fetal but not the adult stomach. In the adult stomach, M-PLA2 was retained only in gastric neck cells, which are known to be the proliferating cells of the gastric mucosa. In a separateinvestigation, we observed strongimmunoreactivity for M-PLA2 in the regenerating cell zone that develops subsequent to mucosal damage, and in intestinal metaplasia of crypt cells in the stomach. Furthermore, abundant expression of M-PLA2 was observed in poorly differentiated adenocarcinoma tissues, particularly in invasive gastric cancer (data not shown). These observations suggest that the expression of M-PLA2 is related to cell growth andlor differentiation. The present study revealed that M-PLA2 immunoreactivity is distributed throughout digestive organs. The distribution and intracellular localization of M-PLA2 in these organs suggests that M-PLA2 may have functions other than regulation of intracellular metabolism.
Acknowledgment We than&Dr U e y aforproviding adult autopsy andfetal necropsy specimens for this study.
Literature Cited
Discussion Membrane-associated-typeGroup I1 PLA2 is hypothesized to regulate cellular phospholipid metabolism and release of prostaglandin (PG) precursors (10). Secretory type Group I1 PLA2 is also secreted into intra-articular fluid ( 5 ) , and a PLA2 in rheumatoid arthritis synovial fluid has been partially sequenced (11). However, the biological function of this form is not clear. Recently, the com-
1. Vadas P, Pruzanski W. Biology of disease: role of secretory phospholi-
pase Az in the pathobiology of disease. Lab Invest 1986;55:391 2. Heinrikson RI., Krueger ET, Keim PS. Amino acid sequence of phospholipase A2-a from the venom of Crotalus adamanteus: a new classification of phospholipase A2 based upon structural determinants. J Biol Chem 1977;212:4913 3. Tojo H, Ono T, Kuramitsu S, Kagamiyama H, Okamoto M. A phos-
Downloaded from jhc.sagepub.com by guest on April 16, 2015
KIYOHARA, EGAMI, SHIBATA, MURATA, OHSHIMA, OGAWA
1664
pholipase A2 in the supernatant fraction of rat spleen: its similarity to rat pancreatic phospholipase A2. J Biol Chem 1988;263:5724 4. Ono T, Tojo H, Kuramitsu S, Kagamiyama H, Okamoto M. Purification and characterizationof a membrane-associated phospholipase A2 from rat spleen: its comparison with a cytosolic phospholipase A2 S-1. J Biol Chem 1988;263:5732 Purzanski W, Vadas P, Stefanski E, Urowitz MB. PhospholipaseA2 activity in sera and synovial fluids in rheumatoid arthritis and osteoarthritis: its possible role as a proinflammatory enzyme. J Rheumatol 1985;12:211 Matsuda Y, Ogawa M, Sakamoto K, Yamashita S, Kanda A. Kohno M, Yoshida N, Nishijima J, Murata A, Mori T. Development of a radioimmunoassay for human group I1 phospholipase A2 and demonstration of postoperative elevation. Enzyme, in press Kohler G, Milstein C. Continuous cultures of fused cells secreting antibody of predefined specificity. Nature 1975;256:495 Hsu SM, &ne L, Fanger H. The use of avidin-biotin-peroxidase complex (ABC) in immunoperoxidase techniques: a comparison between ABC and unlabeled antibody (PAP)procedures.J Histochem Cytochem 1982;29:577
9 Wiyama Y,Egami H, Pout PM. Expression of human tumor-associated antigens in pancreatic cancer induced in Syrian hamsters. Am J Pathol 1990;136:707 10 Van den Bosch H. Intracellular phospholipase A. Biochim Biophys Acta 1980;604:191 11. Vadas P, Stefanski E, Purzanski W. Characterizationof extracellularphospholipase A2 in rheumatoid synovial fluid. Life Sci 1985;36:579 12 Kanda A, Ono T, Yoshida N, Tojo H, Okamota M. The primary structure of a membrane-associatedphospholipase A2 from human spleen. Biochem Biophys Res Commun 1989;163:42 13 Erlandsen SL, Chase DG. Paneth cell function: phagocytosis and intracellular digestion of intestinal microorganisms. I. Hexamita murk J Ultrastruct Res 1972;41:296 14 Erlandsen SL, Chase DG. Paneth cell function: phagocytosis and intracellular digestion of intestinal microorganisms. 11. Spiral microorganism. J Ultrastruct Res 1972;41:319 15 Chipman DM, Sharon N. Mechanism of lysozyme action. Science 1966;19:540
Downloaded from jhc.sagepub.com by guest on April 16, 2015
