GASTROENTEROLOGY

1990;98:607-614

Monoclonal Antibodies to Human Intrinsic Factor ADAM

SMOLKA

and ROBERT M. DONALDSON,

Jr.

Department of Anatomy and Cell Biology, Medical University of South Carolina, Charleston, South Carolina; Department of Internal Medicine, Yale University, New Haven, Connecticut

Mice were immunized with human intrinsic factor, and their lymph node cells were fused with a my eloma cell line hy standard hyhridoma techniques. Eleven of the resulting 227 hyhridomas secreted immunoglobulin G capable of binding to intrinsic factor-cohalamin complex. Cloning by limiting dilution gave 6 clones secreting anti-intrinsic factor antibodies that bound -human intrinsic factor-cobalamin complex with affinities of 13-116 nM; 3 antibodies also bound rabbit intrinsic factor-cobalamin complex. Five antibodies inhibited to some degree the binding of cobalamin by intrinsic factor, and 2 also prevented attachment of intrinsic factor-cobalamin complex to guinea pig ileal receptors. Anti-rabbit intrinsic factor antibodies specifically precipitated a peptide of molecular weight 53,000, corresponding to the molecular weight of rabbit intrinsic factor from homogenates of rabbit gastric mucosal explants biosynthetically labeled with [%]methionine and from culture medium in which the explants were incubated. Indirect fluorescence immunocytochemistry with the antibodies in human and rabbit gastric mucosal sections showed intense selective staining of parietal cells. These results (a) document species differences between human and rabbit intrinsic factors not previously demonstrable with polyclonal anti-intrinsic factor sera; (b) confirm earlier evidence that cobalamin binding and receptor functions occur at separate sites in intrinsic factor; and (c) provide a useful approach to studying structurefunction relations of the intrinsic function molecule.

I

n 1929, Castle defined the principal defect in pernicious anemia as a lack of intrinsic factor (IF) in gastric juice (1). Subsequent studies have delineated the role of IF in intestinal. absorption of the essential vitamin cobalamin (Cbl, Vitamin B,,), as reviewed elsewhere (2). Immunological techniques have been prominent in the study of IF production by gastric mucosa. Hoedemaeker et al. (81 used a y-globulin

fraction of serum with anti-IF activity from pernicious anemia patients to block the uptake of [57Co]-labeled cobalamin ([57Co]-Cbl) by human gastric parietal cells and rat chief cells and thus provided evidence that these cells contained IF. Immunofluorescent studies by Fisher and Taylor in 1969 confirmed that IF was localized in parietal cells in the human gastric mucosa (4). Levine et al. (5) used a monospecific anti-human IF antiserum and peroxidase immunoelectron microscopy to study normal human gastric biopsies. Intrinsic factor was found on membranes of the perinuclear envelope, rough endoplasmic reticulum, Golgi apparatus, and tubulovesicles of resting parietal cells. Pentagastrin stimulation was followed by appearance of IF on canalicular membranes (6). Immunoaffinity purification of IF was reported by Rothenberg in 1965 (7), although B,,-agarose affinity chromatography has subsequently become the standard IF preparative method. More recently, Shepherd et al. (8) purified human IF-Cbl complex from gastric juice by immunoadsorption to agarose-protein A with covalently linked antiIF antibody from a pernicious anemia patient. In an effort to improve the methodology for examining structure-function relations of IF and delineating the mechanism of IF-mediated transport of cobalamin across ileal cells, we have generated a panel of monoclonal antibodies against human IF. We present here an initial characterization of these antibodies by liquid-phase binding assays, immunofluorescence in gastric sections, immunoblot assays, and immune precipitation from gastric mucosal explants. Immunocytochemistry with these antibodies confirms localiza-

Abbreviations used in this paper: Cbl, cnhalamis; EDTA, ethylenediaminetetraacetic acid; PITG fluorescein isothioiqanate; H,KATPase, H,K-adenosine triphosphatase; IF, intrinsic factor; M,, relative molecular weight; PBS, phosphate-buffered saline; SDSPAGE, sodium dodecyl sulfate-polyacrylamide gel electrophoresis; TBS, Tris-htiered saline. 0 3~90hy the American Gastroenterological Association 0026-5085/90/$3.00

608

SMOLKA AND DONALDSON

GASTROENTEROLOGY Vol. 98, No. 3

tion of IF in the parietal cells of human and rabbit gastric mucosa, and immune precipitation of a biosynthetically labeled peptide of relative molecular weight (M,) 53,000 from gastric microsomes and explant culture medium confirms synthesis of IF by this tissue. Materials and Methods Isolation

of Intrinsic

Factor

Intrinsic factor was prepared from neutralized human gastric juice, as previously described [9), and was isolated by affinity chromatography according to the method used by Allen and Mehlman (lo] except that the monocarboxylic acid derivative of cyanocobalamin was linked to agarose (AffigellO.2; Bio-Rad Laboratories, Richmond, Calif.], yielding an affinity ligand with a binding capacity of 100 ng (1.9 pmol] of IF per mg of cobalamin-agarose. The purified IF bound 31 bg cobalamin per mg and gave a single band by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE] [ll] with an M, of 54,000. Monoclonal

Antibodies

One hundred-microgram aliquots of purified IF were injected into the hind footpads of 6-wk-old female BALB/c mice at 3-day intervals: the antigen was emulsified in Freund’s adjuvant for the first 2 injections and diluted in phosphate-buffered saline [PBS) for the remaining 3. Two days after the last injection, inguinal, popliteal, and axial lymph node cells from the immunized mice were fused with the nonsecreting myeloma cell line P3.X63-Ag8.653 as described previously (12). After 18 days’ growth in hypoxanthine-aminopterin-thymidine selective medium (131, secreting hybridomas were identified by dot-blot analysis (14) and their antibodies were screened for anti-IF activity by direct precipitation of [57Co]-Cbl-IF complexes prepared from human gastric juice (151. Positive hybridomas were cloned by limiting dilution, and actively secreting hybridoma clones were subsequently expanded as ascites in pristane-primed syngeneic mice. Monoclonal antibodies were purified from ascites by ammonium sulfate precipitation and protein A-affinity chromatography. Cobalamin-Binding

inhibition

complex

Inhibition

Immunocytochemistry Human gastric biopsies and rabbit and hog gastric mucosal samples fixed in Bouin’s solution were embedded in paraffin, sectioned at 5 pm, cleared in xylene and ethanol, and incubated with monoclonal antibodies diluted 1:50 in 1% bovine serum albumin in PBS for 1 h. The sections were washed 3 times in PBS and incubated for 1 h with 50 ~1 of 1:lOO dilution of fluorescein isothiocyanate (FITCj-conjugated goat anti-mouse immunoglobulin (H + L chain] (Litton Bionetics, Kensington, Md.]. After repeated washing in PBS, the sections were examined by fluorescence microscopy (Leitz photomicroscope). Some hog gastric musosal sections received horseradish peroxidase-conjugated goat anti-mouse immunoglobulin (1:lOO dilution] instead of FITC second antibody, followed by incubation in 0.05% 4-chloronaphthol, 0.015% H,O, in PBS, with visualization by light microscopy.

Western

Blotting

Purified human IF was resolved by SDS-PAGE (10% acrylamide), and the resulting peptide pattern was transferred to nitrocellulose sheets as described (18). The sheets were air-dried and incubated for 1 h with 3% gelatin in 20 mM Tris and 150 mM NaCl, pH 7.5 (Tris-buffered saline, TBS] then for 2 h with 1:2000 dilutions of monoclonal antibodies in 1% gelatin in 1% Tween ZO/TBS (TTBS). The sheets were washed 3 times in TTBS and incubated for 1 h with 1:2500 dilution in TI’BS of goat anti-mouse immunoglobulin conjugated to horseradish peroxidase (Bio-Rad Laboratories]. After washing as above, the color reaction was developed by incubation in 0.05% 4-chloronaphthol and 0.015% H,O, in TBS.

Assay

This assay is based on the observation that anti-IF antibody inhibits the binding of [5’Co]-Cbl by IF and was carried out as described previously (15,16). Briefly, hybridoma supernatants were incubated with an IF-rich preparation of gastric juice 19)for 10 min at 25°C. [57Col_Cbl was added, and after 20 min incubation, free [57Co]-Cbl was removed with charcoal. Bound radioactivity remaining in the clear supernatant was measured. Reagent blanks and a hybridoma supernatant unable to precipitate IF served as controls. lleal Attachment

measures the ability of anti-IF antibodies to inhibit this process. The assay was performed as described previously (17). Briefly, guinea pig brush border membrane preparations were incubated for 30 min with [57Co]-Cbl-IF complex that had previously been incubated with anti-IF monoclonal antibodies, reagent blanks, or inactive monoclonal antibody. Brush border membranes were then separated from suspension by centrifugation, washed twice, and assayed for radioactivity.

Assay

This assay, based on the uptake qf IF-[57Co]-Cbl by guinea pig ileal brush border membranes,

Biosynthetic Proteins

Labeling

of Gastric

Mucosal

Gastric mucosal explants 6 mm in diameter were peeled from the underlying muscularis; placed into 0.5 ml methionine-free RPM1 1640 culture medium with 10% dialyzed fetal calf serum, 100 U/ml penicillin, 100 rg/ml streptomycin, and 200 PCi [35S]-methionine (1100 Ci/mmol); and incubated at 37°C in 5% CO,/95% air for 6 h. Explants were disaggregated in 150 mM NaCI, 10 mM Tris-HCI, pH 7.5, and 2 mM ethylenediaminetetraacetic acid (EDTA) (NET] using a Brinkman Polytron (Westbury, N.Y.) for 60 sat setting 5 and centrifuged (Sorvall RC60; Du Pant, Wilming ton, Del.) at 12,000 x g for 45 min. The supernatant was recentrifuged at 100,000 x g for 1 h, yielding a microsomal

INTRINSIC

March 1990

FACTOR

MONOCLONAL

ANTIBODIES

609

pellet. Aliquots of the homogenate, microsomal pellet, and culture medium were solubilized by addition of Nonidet P40 (NP40; Sigma, St. Louis, MO.) to l%, and insoluble material was removed by 5 min of centrifugation at 15,000 x g (Beckman Microfuge. Fullerton, Calif.]. Immune Precipitation Twenty microliters of heat-killed, formalin-fixed Staphylococcus aureus bacteria (Immuno-Precipitin; BRL, Gaithersburg, Md. 10% wt/vol in NET) was incubated with 0.5-m] aliquots of solubilized cell fractions for 1 h at 4’C and then removed by 30 s of centrifugation in the Microfuge. Five micrograms of anti-IF monoclonal antibody was added, and after 1 h at 4”C, 20 ~1 bacteria was added for 1 h at 4’C. The bacteria were recovered by centrifugation and washed 3 times in NET/l% NP40, once in 0.5 M NaCl in NET, and once in distilled water. The bacteria were suspended in Laemmli sample buffer (111, and the solubilized immune precipitates were analyzed by SDS-PAGE fluorography (10% acrylamide).

Fusion of lymph node cells from mice immunized with human IF with a myeloma cell line resulted in the formation of 227 hybridoma clones, 162 of which were found to secrete measureable levels of immunoglobulin by dot-blot analysis. Culture supernatants from these hybridomas were tested for anti-IF activity by direct precipitation of [“Co]-Cbl-IF complexes from human gastric juice (15). Eleven supernatants were found to bind to the complex, and their parent hybridomas were cloned by limiting dilution. Six of the resulting clones preserved their IF-binding activity and were expanded as ascites in syngeneic mice. The 6 secreted monoclonal antibodies bound to human IF-Cbl complex with affinities ranging from 13-116 nM (Table 1). Three of the antibodies (IF 1.18, IF 5.6, and IF 11.5) also bound to rabbit IF-Cbl complex.

Table 1. Binding Capacity and Affinity Constants of Six Monoclonal Antibodies Against Human Intrinsic Factor Bindingcapacity” IF 1.9 IF 1.16 IF 5.6 IF 6.7 IF 11.5 IF 13.20 ~3.16~

1.18

5.6

8.7

11.5

0 13.20

polyclonzl

Figure 1. Effects of anti-IF monoclonal antibodies and one anti-IF serum from a pernicious anemia patient (polyclonal) on the binding of Cbl to human IF and on the binding of IF-Cbl complex to ileal receptor on guinea pig brush border membranes.

Results

Antibody

1.9

(clg IF/mg IgGl 0.47 2.64 5.13 0.15 0.77 0.37 0

Affinity constant (WnM 1 84 20 16 116 13 45 -

IgG. immunoglobulin G. “Measured by immune precipitation (cf. reference 15). bSupernatant from a hybridoma (same fusion) secreting nonspecific immunoglobulin G.

Preincubation of human gastric juice with monoclonal antibodies followed by addition of [57Co]-Cbl and subsequent quantitation of the IF-Cbl complexes showed that 5 antibodies inhibited the formation of the complex to varying extents, as shown in Figure 1, with a maximal inhibition of 70% in the case of IF 5.6. A nonspecific monoclonal immunoglobulin G from the same fusion, showing no anti-IF activity and used as a control antibody in these studies, showed no inhibition of formation of the complex. Figure 1 also depicts the effects of pre-incubation of the 6 antibodies with IF-Cbl complex on the subsequent attachment of the complex to guinea pig brush border membranes. Three antibodies (IF 1.9, IF 1.18, and IF 13.20) did not affect this attachment; IF 5.6 and IF 11.5 inhibited binding of the complex to its ileal receptor by 33% and 40%, respectively. Included in Figure 1 are the corresponding data for a previously described polyclonal serum obtained from a patient with pernicious anemia (19). This polyclonal serum inhibited both the binding of Cbl to IF and the attachment of IF-Cbl complex to ileal receptors. Binding of antibodies to human IF was assessed by immunoblotting, as shown in Figure 2. Highly purified human IF was resolved by SDS-PAGE on a 10% acrylamide gel and was found by Coomassie blue staining to consist of a single polypeptide band of M, 54,000 (lane a]. This peptide pattern was transferred electrophoretically to a nitrocellulose sheet and probed with the monoclonal antibodies. In two separate immunoblotting experiments, 6 anti-IF antibodies (IF 1.9, IF 1.18, IF 5.6, IF 8.7, IF 11.5, and IF 13.20) bound to a

610

SMOLKA

a

AND DONALDSON

bcdefgh

GASTROENTEROLOGY

Vol. 98, No. 3

i

94

67

43

30

single peptide corresponding to M, 54,000 [lanes b-f]. Substitution of a nonspecific monoclonal immunoglobulin G for anti-IF antibodies when probing the nitrocellulose blot eliminated the signal (lane h). Three different monoclonal antibodies against hog gastric H, K-adenosine triphosphatase (H,K-ATPase) (20) showed no reactivity with human IF by immunoblotting; this is shown for one antibody (HK 1.21) in lane i of Figure 2. The cellular localization of the epitope(s) recognized by the anti-IF monoclonal antibodies was sought by immunocytochemistry at the light microscopic level. As detailed above, human, rabbit, and hog gastric mucosal sections (5 pm thick] were incubated with monoclonal antibody, washed, and then incubated with FITC-conjugated second antibody. Antibodies IF 5.6 and IF 11.5 stained human, rabbit, and hog tissue, IF 1.18 stained only human biopsy tissue, and IF 13.20 stained only rabbit tissue. In all species, the only cell type stained was parietal. Figure 3A shows IF 5.6 staining in rabbit tissue, and Figure 3B shows IF 1.18 staining in human biopsy tissue. The distribution and intensity of staining was comparable to that found previously in hog and rabbit gastric mucosa with anti-H,K-ATPase monoclonal antibodies (20): anti-IF antibody was localized in the cytoplasm of parietal cells, with no nuclear staining. No antibody staining was evident in the chief cells, mucous neck cells, surface epithelial cells, or enterochromaffin cells, nor was it apparent in myoepithelial cells, plasma cells, or the interstitial supporting matrix of the mucosa. Basolateral membranes of cells lining the gastric glands were not stained. The staining was eliminated by prior incubation of the monoclonal antibody with homogeneous human IF and by substitution of a nonspecific

Figure 2. Immunoblot analysis of anti-IF monoclonal antibodies using puri5ed human IF as the bound antigen (lone a, Coomassie Blue stainimp). Lanes b-g. IF 1.9, IF 1.18, IF 5.6, IF 8.7, IF 11.5, and IF 13.20, respectively. Lane h. Nonspeci5c mouse monoclonal immunoglobulin G. Lone i. Anti-H,KATPase monoclonal antibody HK 1.21. The position of molecular weight standards is shown at left in kilodaltons.

mouse monoclonal immunoglobulin G for the anti-IF antibodies in the first step of the procedure. Immunocytochemical treatment of hog fundic gastric mucosal sections documented intense staining of parietal cells by monoclonal anti-IF antibodies IF 11.5 and IF 5.6: this is shown for IF 5.6 in Figure 4A. To confirm that hog parietal cells were being targeted by these antibodies, we sought evidence for the coexistence of IF and H,K-ATPase epitopes within the same cells. Hog gastric mucosal sections were stained with IF 5.6, photographed, and then treated to remove both the first (monoclonal) and second (FITC-labeled) antibodies (21). No fluorescence was visible on the sections after this treatment. The sections were restained with a monoclonal antibody (HK 1.21) directed against the gastric H,K-ATPase responsible for acidification of gastric juice (22) and rephotographed. Subsequent comparison of photographs (Figures 4A and 4B) revealed that every cell staining positively with anti-IF antibody was also positive with anti-H,K-ATPase antibody. Parietal-cell IF synthesis and processing were examined with monoclonal anti-IF antibodies during culture of rabbit gastric mucosal explants. Explants were incubated for 6 h with [35S]-methionine then homogenized, and microsomal fractions were prepared by differential centrifugation. After the fractions were solubilized in NP40, the microsomal fraction and the explant culture medium were subjected to immune precipitation by antibody IF 5.6, and the results were analyzed by SDS-PAGE fluorography, as shown in Figure 5. The microsomal fraction (lane a) shows incorporation of [35S]-methionine into many polypeptides with a wide range of molecular weights, the most

March 1990

INTRINSIC FACTOR MONOCLONAL

ANTIBODIES

611

a proteolytic fragment of secreted IF: the source of the 62-kilodalton band is unknown. No bands were apparent on the fluorograms when nonspecific mouse monoclonal immunoglobulin G replaced IF 5.6 in immune precipitation (lane b, Figure 5). Discussion These results document the production and partial characterization of a panel of 6 monoclonal antibodies against human IF. Three antibodies (IF 1.9, IF 1.18, and IF 13.20) inhibited Cbl binding to IF but had no effect on binding of preformed IF-Cbl to the ileal receptor. In contrast, 2 other antibodies (IF 5.6 and IF 11.5) inhibited both of these binding activities

Figure 3. Immunocytochemistry of gastric mucosai sections with anti-IF monoclonal antibodies. Cell labeling was visualized with FPIC-conjugated second antibody under fluorescent lllumhation. A. Rabbit tissue stained with antibody IF 5.6 (original magnification x400). B. Human biopsy stained with antibody IF 1.18(original magnification x 160).

prominent being two bands with molecular weights of approximately 43,000 (pepsinogen) and 58,000 (unknown). Immune precipitation of this material with IF 5.6 showed only a single polypeptide with M, 53,000, the molecular weight of native rabbit IF (lane c). Immune precipitation of the explant culture medium with IF 5.6 (lane d) gave a prominent band at 53 kilodaltons, representing soluble IF secreted from the gastric mucosal samples during the labeling phase, and two faint bands at 62 and 22 kilodaltons. It is unlikely that these latter bands are nonspecific contaminants of precipitated S. aureus immune complexes because the bands were absent when IF 5.6 preabsorbed with human IF was used for immune precipitation (lane e, Figure 5). The 22-kilodalton band could be

Figure 4. lmbect immlulofluoresceu cemhxogqhshowingsequential staining of a single hog gastric mucosal section with anti-IF and anti-H,K-ATPase monoclonal antibodies. A. Antibody IF 5.6. B. Same section restained with HK 1.21

612

SMOLKA

AND DONALDSON

abc

GASTROENTEROLOGY

de

67-

43-

21Figure

5. SDS-PAGE fIuorography of immune precipitates recovered from hiosyntheticahy labeled rabbit gastric microsomes and gastric explant culture medium. Lone a. Gastric microsomes. Lane b. Precipitate from microsomes using nonspechlc mouse monoclonal hnmunoglohulin G. Lane c. Precipitate from microsomes with IF 5.6. Lane d. Precipitate from explant culture medium with IF 5.6. Lane e. Precipitate from explant culture medium with IF 5.6 preabsorbed with human IF. Position of molecular weight standards is shown at left in kilodaltons.

(Figure 1). Polyclonal anti-IF antibodies in the sera of patients suffering from pernicious anemia consistently show type I specificity, blocking formation of an IF-Cbl complex (15,16). Some of these sera also exhibit type II specificity, binding to the IF-Cbl complex and preventing its attachment to the ileal receptor (23,241. Antibodies IF 5.6 and IF 11.5 clearly show a combination of type I and type II sites, with immunoglobulin G binding at one site conformationally affecting reactivity at the other. Close proximity of type I and type II sites on the IF molecule, such that immunoglobulin G binding to one site directly blocks access to the second, seems to be ruled out because antibodies IF 1.9, IF 1.18, and IF 13.20 show only type I specificity. The mechanism of these differential antibody inhibitions of IF reactivity toward Cbl and the ileal receptor remains unclear, but these data strongly support other evidence that the site on the IF molecule that binds Cbl is separate from that responsible for ileal receptor attachment (21. The localization of monoclonal anti-IF antibody staining to parietal cells of human and rabbit gastric mucosae is consistent with previous immunocytochemical and biochemical studies (3-6,25) identifying gas-

Vol. 98, No. 3

tric parietal cells as the source of IF in these species. Positive staining of hog parietal cells with 2 of the antibodies was unexpected because pyloroduodenal mucus cells have long been associated with IF secretion in the hog (3). The colocalization of IF and H,K-ATPase monoclonal antibodies to the same cell type in hog fundic gastric mucosa argues strongly for a parietal-cell role in IF secretion in the hog, Recently, Lee et al. (26) showed that monospecific antisera against rat IF stained parietal cells in hog stomach, Positive staining was confirmed by its reversal with rat IF purified to homogeneity. Further studies of tissue distribution of IF secretion in the hog are obviously needed. The high intensity of staining and the relative absence of background make monoclonal antibodies well suited to morphologic studies of IF localization and intracellular distribution by electron microscopy. Particularly important is the question of whether de novo synthesized IF is inserted into the same tubulovesicular membrane compartments that contain the proton-pumping H,K-ATPase responsible for acid secretion. Using immunoelectron microscopy, Levine et al. found IF on membranes of the perinuclear envelope, rough endoplasmic reticulum, Golgi, and tubulovesicles of human parietal cells (5). Shortly after pentagastrin stimulation, however, IF-staining tubulovesicles were found close to secretory canalicular membranes; 15 min after stimulation, IF was bound to canalicular membranes, and after 1 h 25% of the parietal cells had regained a resting morphology, with IF once again on basal rough endoplasmic reticulum and tubulovesicular membranes (6). Similarly, Smolka et al. (22) localized H,K-ATPase to tubulovesicles in unstimulated rabbit parietal cells, with little or no labeling of rudimentary secretory canalicular membranes by monoclonal antibody. In stimulated parietal cells, however, the antibody labeled secretory canalicular membranes most intensely. These electron microscopic data on IF and H,K-ATPase distribution are consistent with the concept that tubulovesicles fuse with canalicular membrane upon stimulation (27). However, the data do not exclude the existence of 2 populations of tubulovesicles, one of which accumulates IF while the other has acid secretory potential by virtue of its integral membrane H,K-ATPase. Fusion of both tubulovesicular populations upon stimulation would then result in cosecretion of IF and protons (28). Studies of colocalization of IF and H,K-ATPase antibodies at the electron microscopic level using immunogold technique may clarify this issue. The precipitation by antibody IF 5.6 of a peptide of M, 53,000 from a microsomal fraction of rabbit gastric mucosal explants is consistent with close association of IF with microsomal membranes at some stage in its synthesis and processing by rabbit parietal cells. In

INTRINSIC FACTOR MONOCLONAL

March 1990

contrast, Serfilippi and Donaldson (19) found that an anti-IF antibody prepared from the serum of a patient with pernicious anemia precipitated at least 4 peptides from rabbit gastric tissue extracts. Because precipitation in that study occurred from supernatants of desoxycholate-solubilized tissue, with no intact microsomes present, the multiple coprecipitating bands probably represent tissue antigens recognized by nonIF specificities present in the polyclonal antibody. In the present study, the same anti-IF monoclonal antibody (IF 5.6) precipitated a de novo synthesized 53kilodalton peptide from culture medium bathing the rabbit gastric explants. These results emphasize the potential usefulness of anti-IF monoclonal antibodies in the study of synthesis and secretion of IF. Although the complete primary structure of human IF is unknown, the sequence of 5 peptide fragments of human IF, including the amino terminal sequence, has been reported (29). Comparison of these sequences with the primary structure of rat gastric H,K-ATPase (30) shows an identical pentapeptide (glycine-leucine-valine-valine-serine, residues 265269] in the ATPase. The primary structure of rat IF, deduced from a cyclic deoxyribonucleic acid clone (31), contains a homologous pentapeptide (glycineleucine-isoleucine-valine-serine, residues 371-375). If H,K-ATPase and IF are both sequestered in the same tubulovesicular membranes, as immunocytochemical evidence seems to imply, it is possible that they share certain common features that direct the insertion of the nascent peptides into this membrane compartment and no other. The antibodies described here will be useful in studies of cell-free translation by gastric messenger RNA as an approach to understanding the synthetic and processing steps undergone by nascent IF peptide. Confirmation that IF and H,K-ATPase reside in the same tubulovesicular membrane may emerge from biochemical and immunochemical analyses of native gastric microsomal membranes immune precipitated by the antibodies; whether the relative microsomal distribution of these 2 peptides is modulated as a function of the secretory status of the parietal cell is of great interest and may also be approached by immunogold electron microscopy with anti-IF and anti-H,KATPase monoclonal antibodies.

References 1. Castle WB. Observations on the etiologic relationship of achylia gastrica to pernicious anemia. I. Effect of administration to patients with pernicious anemia of contents of normal human stomach recovered after ingestion of beef muscle. Am J Med Sci 1929;78:748-64. 2. Donaldson RM Jr. Intrinsic factor and the transport of cobalamin. In: Johnson LR, ed. Physiology of the gastrointestinal tract. New York: Raven, 1987;959-73.

ANTIBODIES

613

3. Hoedemaeker PJ. Abels J. Wachters JJ, Arends A, Nieweg HO. Investigations about the site of production of Castle’s gastric intrinsic factor. Lab Invest 1964;13:1394-9. 4. Fisher JM, Taylor KB. The intracellular localization of Castle’s intrinsic factor by an immunofluorescent technique using antoantibodies. Immunology 1969:16:779-84. 5. Levine JS, Nakane PK. Allen RH. Immunocytochemical localization of human intrinsic factor: The non-stimulated stomach. Gastroenterology 1980;79;493-502. 6. Levine JS, Nakane PK. Allen RH. Human intrinsic factor secretion: immunocytochemical demonstration of membraneassociated vesicular transport in parietal cells. J Cell Biol 1981;90:644-55. 7 Rothenberg SP. Immunologic isolation of human intrinsic factor. Blood 1965;26:868. 8. Shepherd HA, Priddle JD, Jenkins WJ, Jewel1 DP. The preparation of human intrinsic factor-cobalamin complex from human gastric juice by immunoadsorption. Clin Chim Acta 1984;139:15565. 9 Kapadia CR, Serfilippi D. Voloshin K, Donaldson RM. Intrinsic factor-mediated absorption of cobalamin by guinea pig ileal cells. J Clin Invest 1983;7:440-8. 10 Allen RH, Mehlman CS. Isolation of gastric vitamin B,,-binding proteins using affinity chromatography. I. Purification and properties of human intrinsic factor. J Biol Chem 1973;248:3660-9. 11 Laemmli UK. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 1970;227:680-4. 12. Kearney JF, Radbruch A, Leisegang B, Rajewsky K. A new mouse myeloma cell line that has lost immunoglobulin expression but permits the construction of antibody-secreting cell lines. J Immunoll979:123:1548-50. 13. Littlefield JW. Selection of hybrids from matings of fibroblasts in vitro and their presumed recombinants. Science 1964;145:70910. 14.Hawkes R, Niday E, Gordon J. A dot-immunobinding assay for monoclonal and other antibodies. Anal Biochem 1982;119: 142-7. 15.Schade SG, Abels J. Schilling RF. Studies on antibody to intrinsic factor. J Clin Invest 1967;46:615-20. 16.Ardeman S, Chanarin I. A method for the assay of human gastric intrinsic factor and for the detection and titration of antibodies against intrinsic factor. Lancet 1963;2:1350-4. 17.Mathan VI. Babior BM, Donaldson RM. Kinetics of the attachment of intrinsic factor-bound cobamides to ileal receptors. J Clin Invest 1974;54:598-608. 18.Burnette WN. Western blotting. Electrophoretic transfer of proteins from sodium dodecyl sulfate polyacrylamide gels to unmodified nitrocellulose and radiographic detection with antibody and radioiodinated protein A. Anal Biochem 1981;112:195203. 19.Serfilippi D, Donaldson RM Jr. Production and secretion of intrinsic factor by isolated rabbit gastric mucosa. Am J Physiol 1986;251:G287-92. 20. Smolka A, Weinstein WM. Immunoassay of pig and human gastric proton pump. Gastroenterology 1986;90:532-9. 21.Tramu G, Pillez A, Leonardelli J. An efficient method of antibody elution for the successive or simultaneous localization of two antigens by immunocytochemistry. J Histochem Cytochem 1978;26:322-4. 22. Smolka A, Helander HF, Sachs G. Monoclonal antibodies against gastric proton potassium ATPase. Am J Physiol1983:245: G589-96. 23. Samloff IM, Kleinman MD, Turner M. Sobel V. Jeffries GH. Blocking and binding antibodies to intrinsic factor and parietal cell antibody in pernicious anemia. Gastroenterology 1968:55: 575-83. IL.

614

SMOLKA

AND DONALDSON

24. Schade SG, Feick P, Muckerheide M, Schilling RF. Occurrence in gastric juice of antibody to a complex of intrinsic factor and vitamin B. N Engl J Med 1966;275:528-31. 25. Schepp W, Miederer SE, Ruoff H-J. Intrinsic factor secretion from isolated human gastric mucosal cells. Biochim Biophys Acta 1984;804:192-9. 26. Lee EY, Alpers DH, DeSchryver-Kecskemeti K. Upper G.I. accessory organs produce both intrinsic factor (IF] and R protein. Gastroenterology 1988;94:A254. 27. Forte TM, Machen TE, Forte JG. Ultrastructural changes in oxyntic cells associated with secretory function: membrane recycling hypothesis. Gastroenterology 1977;73:941-55. 28. Jiron C, Roman0 M, Michelangeli F. A study of dynamic membrane phenomena during the gastric secretory cycle: fusion, retrieval and recycling of membranes. J Membr Biol 1984;79:119-34. 29. Nexo E, Olesen H, Hansen MR. Bucher D, Thomsen J. Primary structure of human intrinsic factor; progress report on cyanogen bromide fragmentation. Stand J Clin Lab Invest 1978;38:649-53.

GASTROENTEROLOGY

Vol. 98, No. 3

30. Shull GE, Lingrel JB. Molecular cloning of the rat stomach (H+ + K+)-ATPase. J Biol Chem 1988:261:16788-91. 31. Dieckegraefe BK, Seetharam B, Banaszak L, Leykam JF, Alpers DH. Isolation and structural characterization of a cDNA clone encoding rat gastric intrinsic factor. Proc Nat1 Acad Sci 1988;85: 46-50.

Received April 24,1989. Accepted August 21,1989. Address requests for reprints to: Adam Smolka, Ph.D., Department of Anatomy and Cell Biology, Medical University of South Carolina, 171 Ashley Avenue, Charleston, South Carolina 29425. This work was supported in part by National Institutes of Health Grant AM 34092 (AS.). The authors thank Diane Zettel for skilled technical assistance and Dr. George Sachs for provision of laboratory facilities at the Center for Ulcer Research and Education.

Monoclonal antibodies to human intrinsic factor.

Mice were immunized with human intrinsic factor, and their lymph node cells were fused with a myeloma cell line by standard hybridoma techniques. Elev...
1MB Sizes 0 Downloads 0 Views