Isolation and Characterization of the Ileal Receptor for Intrinsic Factor-Cobalamin*

The receptor for intrinsic factor-cobalamin (vitamin BIZ) has been purified 65,000-fold from 2.5 kg of canine ileal mucosa with a recovery of 26%. The initial purifi- cation steps involved solubilization of ileal mucosal homogenates with Triton X-100, followed by ethanol precipitation and dialysis. The soluble receptor was then saturated with hog intrinsic factor-cobalamin and the receptor-hog intrinsic factor-cobalamin complex was adsorbed to anti-hog intrinsic factor-Sepharose. After extensive washing, the receptor was eluted with 5 m~ EDTA at pH 5.0, leaving behind the hog intrinsic factor-cobalamin that remained bound to the anti-hog intrinsic factor-Sepharose. Gel filtration on Bio-Gel A-5m with 0.1% Triton X-100 gave two peaks of receptor activity with apparent mo- lecular weights of 7,500,000 and 5,000,000 indicating that the receptor aggregates under these conditions. Polyacrylamide disc gel electrophoresis with 0.1% Triton X-100 gave a single protein band that barely en- tered 4% gels, did not stain for carbohydrate, and co-incided with the presence of intrinsic factor-cobalamin-binding ability. Polyacrylamide gel electrophoresis with sodium dodecyl sulfate gave one major band with an apparent molecular weight of

' The abbreviat.ion used is: Cbl, cobalamin; SDS, sodium dodecyl sulfate. is bound predominantly by R protein' in gastric juice and becomes bound to intrinsic factor after the R protein moiety is partially degraded by pancreatic proteases in the jejunum (4-6). The intrinsic factor-Cbl complex then binds to receptors that are present on ileal brush borders, but not on jejunal brush borders (7)(8)(9)(10)(11). These receptors are specific for the intrinsic factor-Cbl complex and do not bind free Cbl, free intrinsic factor, or other protein-Cbl complexes (8). The binding of intrinsic factor-Cbl to these receptors markedly facilitates the absorption of Cbl although little is known about. the subsequent steps in the absorptive process (12).
Ileal receptors for intrinsic factor-Cbl have been solubilized and partially purified but their characterization has been hindered because of the extremely small amount of receptor present in the ileum (4, 12-16). We have shown previously (8) that canine ileal mucosa has a somewhat higher number of intrinsic factor-Cbl receptors than most other species. We now report the successful purification of the intrinsic factor-Cbl receptor from this source using a newly developed immunoaffinity chromatography technique, This technique utilizes rabbit anti-hog intrinsic factor-Sepharose to bind the receptorhog intrinsic factor-Cbl complex with the subsequent elution of the receptor alone, at pH 5.0, in the presence of EDTA. The receptor thus obtained is highly pure and this has enabled us to characterize a number of its physical properties.

EXPERIMENTAL PROCEDURES
Materials CN-["Co]Cbl and Na-"'"l were obtained from Amersham Corp., Arlington Heights, 1L. Sepharose-4B, horse spleen ferritin, Escherichia coli B-galactosidase, beef liver catalase, human IgG, and ovalbumin were obtained from Sigma Chemical Co., St. Louis, MO. Bio-Gel P-30, 100-200 mesh, Bio-Gel A-5m, 200-400 mesh, and Enzyrnobeads were obtained from Bio-Rad Laboratories, Richmond, CA. Hog intrinsic factor (171, human intrinsic factor (18), an abnormal human intrinsic factor (19), and hog H protein (17) were obtained as described previously. Canine intrinsic factor was purified from canine stomachs using a procedure similar to that employed for the purification of human intrinsic factor (18). Antibodies to various species of intrinsic factor were raised in rabbits and covalently coupled to Sepharose-4B as described previously (20). The equivalent of 2 ml of antiserum were coupled per ml of Sepharose.

Methods
Receptor Assays-Membrane-bound receptor present in canine ' The term "R protein" was originally devised to denote a cobalamin-binding protein in human gastric juice that was devoid of intrinsic factor activity. It was designated as protein "R" because of its rapid mobility on electrophoresis. Subsequently, immunologically related proteins have been observed in a variety of human tissues and body fluids and have been referred to as the R proteins. ___l__x_I 3785 Ileal Receptor for Intrinsic Factor-Cobalamin ileal homogenates was assayed as described previously (8). Soluble receptor was assayed by a modification of the method of Cotter and Rothenberg (13). Test tubes contained the following in a final volume of 1.0 ml: Tris-HCI, pH 7.5, 10 pmol; NaCI, 140 pmol; CaCL or EDTA, IO pmol; hog intrinsic factor-['CoICbl, 0 to 1.2 pmol; Triton X-100, 0.1% (v/v); and various amounts of soluble receptor. After incubating for 1 h at 22"C, the tubes were cooled to 4"C, and 50 p1 of normal human serum was added to provide carrier protein, followed by the addition of 660 pl of 3.9 M (NH,),SO, that had been adjusted to pH 7.0 with NaOH. The test tubes were blended on a Vortex mixer, centrifuged at 10, OOO X g for 15 min at 4"C, and the supernatant containing unbound hog intrinsic factor-["CoICbl, and the pellet containing receptor-hog intrinsic factor-["'ColCbl were assayed for radioactivity. Specific binding was taken as that obtained with CaC1, minus that obtained with EDTA. K<< was calculated as previously described (8).
Pol.yacrylamide Gel Electrophoresis-Polyacrylamide disc gel electrophoresis was performed on 4% acrylamide gels containing 0.1% Triton X-100 and stained for protein or carbohydrate as described previously (21). Receptor activity was localized using unstained gels that were quickly frozen on dry ice and cut into 1.1-mm slices. Individual gel slices were eluted by placing them in test tubes containing I ml of 5 mM potassium phosphate, pH 7.4, containing 140 mM NaCI, 5 mM KCI, 0.1% Triton X-100, and 50 pg of bovine serum albumin and gently agitated overnight on a platform shaker. The supernatants were then assayed for receptor activity as described above.
SDS-Polyacrylamide Gel Electrophoresis-Polyacrylamide gel electrophoresis in the presence of SDS was performed by the method of Weber and Osborn (22) using 7.5% gels. The molecular weight markers utilized were human IgG (molecular weight, 150,OOO), E. coli B-galactosidase (molecular weight, 13O,OOO), beef liver catalase (molecular weight, 63,OOO), and ovalbumin (molecular weight, 43,000). In a separate gel experiment the receptor was treated with protease inhibitors (phenylmethylsulfonyl fluoride, n-ethylmalamide, and EDTA, all 2.5 mM), before treatment with P-mercaptoethanol and SDS. Triton X-100 was removed from receptor preparations prior to electrophoresis by washing with 30% ethanol in an Amicon concentrator using a UM-10 filter. For 10 ml of solution containing receptor, approximately 100 ml of 30% ethanol, added in three batches, was required to remove >99%, of the Triton X-100 present based on measuring the ALXO in the effluent from the Amicon concentrator. After concentrating to a volume of approximately 100 pl, samples were taken up in 0.1% SDS and subjected to electrophoresis.
Isoelectric Focusing-Isoelectric focusing was performed in tubular gels, 0.5 cm diameter and 20 cm long, as described by OFarrell (23). Ampholines, 2%; acrylamide, 4%; and Triton X-100, 2%; were polymerized in the presence of 4 M urea. Gels were prerun for 15 min a t 100 V and 45 min a t 200 V prior to sample application. Samples containing 20-30 pg of protein, 2R ampholine, and 4 M urea were then subjected to electrophoresis for 6 h at 400 V. Gels were stained for protein with Coomassie brilliant blue; pH values along a simultaneously run unstained gel were determined with a contact microelectrode.
Preparation ofAntibody to the Receptor-Antibody to the purified receptor was raised in New Zealand male rabbits. Purified receptor, 50 pg in a volume of 1 ml was mixed with 1 ml of complete Freund's adjuvant and injected subcutaneously in three different areas on the back. Rabbits were boosted with the same dose on days 7, 14, and 21. Blood was drawn 10 days after the last booster dose and every 7th day thereafter for about 4 weeks. Serum was prepared and pooled and utilized for studies employing anti-receptor antibody.
Protein Determination-Protein was measured by the method of Lowry (25) with a minor modification (26) when Triton X-100 was Amino Acid Analysis-Samples containing approximately 60 pg of purified protein and a measured amount of hog intrinsic factor-Cblbinding ability were dried by lyophilization. The dried samples were dissolved in 1 ml of 6 N HC1, heated at 110°C for 24 h, and dried again. Amino acid analysis was performed using chromatography on that samples were hydrolyzed for 4 h with 4 N HCI at 110°C.
Dionex-D-500. Amino sugars were analyzed in the same way except Purification of the Receptor-The mucosa from the distal halves of canine intestines was scraped off with glass slides and stored at -70°C for up to 2 weeks. Mucosa, 2.5 kg, was thawed, 10 liters of 5 mM potassium phosphate, pH 7.4, containing 140 mM NaCl and 5 mM KC1 were added, followed by homogenization in a Waring blendor for 1 min at top speed at 4°C. Triton X-100 was added to give a final concentration of 196 (v/v), and after stirring for 16 h a t 4"C, the homogenate was centrifuged at 20,000 X g for 30 min. Ethanol stored a t -2O' C was added slowly to the supernatant fraction to give a final concentration of 30% (v/v). After stirring for 1 h a t 4OC the samples were centrifuged at 200,000 X g and the pellets were suspended in 3 liters of the buffer solution described above followed by dialysis against 10 liters of this solution for 18 h with three changes of dialysate. The dialyzed material was then treated with Triton X-100 to give a final concentration of 1% (v/v) and allowed to stir overnight a t 4OC followed by centrifugation for 1 h at 150,000 X g and collection of the supernatant. The ethanol precipitation step was then repeated as just described except that the initial pellets were resuspended in only 500 ml of buffer solution.
Assays of receptor activity following the second ethanol precipitation step usually gave values of 10-15 nmol of total hog intrinsic factor-Cbl-binding ability. Based on the actual value obtained for a given preparation at this stage, a 2-fold excess of hog intrinsic factor-['CoICbl, i.e. 20-30 nmol, was added to the samples which had been adjusted to contain 2.5 mM CaC12 and 1.25 mM MgSO.,. After incubating for 1 h at 2'2"C, the sample was cooled to 4°C and applied to a 2.5 cm diameter and 1 cm tall column of rabbit anti-hog intrinsic factor-Sepharose which had a capacity to bind approximate1.v 40 nmol of hog intrinsic factor-Cbl. The column was washed immediately before the sample was applied with the following solutions at a flow rate of 30 ml/h: wash A, 250 ml of 5 mM potassium phosphate-HCI, pH 5.0, 140 mM NaCI, 5 mM KCI, 5 mM EDTA, and 1% Triton X-100; wash E, 250 ml of potassium phosphate, pH 7.4, 140 mM NaCI, and 1% Triton X-100; and wash C, 250 ml of potassium phosphate, pH 7.4, 140 mM NaCl, 2.5 mM CaC12, 1.25 mM MgSOI, and 1B Triton X-100. The sample was applied at, a flow rate of approximately 20 ml/hr a t 4°C. The column was then washed sequentially with 500 ml of wash C and 500 ml of wash B a t a flow rate of 30 ml/h. The receptor was then eluted with 65 ml of uash A containing 0.1Y Triton X-100 at a flow rate of 50 ml/h at 4°C. The pH was immediately adjusted to 7.4 with 1 M NaOH and 1.0 M CaCl, was added to give a concentration of 5 mM. Receptor was stored at 4-5°C for up to 1 month without loss of activity.  illustrates the (NH&S04 precipitation assay utilized for measuring soluble receptor and demonstrates that the receptor-intrinsic factor-["Co]Cbl complex precipitates at lower concentrations of (NH4)2S04 than does unbound intrinsic factor-["CoICbl. Routine assays utilized 1.5 M (NH,),SO, which precipitated 928 of the receptor-intrinsic factor-["'Co]C!bl complex while precipitating only 3% of the unbound intrinsic factor-["CoICbl. When CaC1, was replaced by EDTA in assays containing receptor and intrinsic factor-["CoICbl, the amount of intrinsic factor-["]Cbl precipitated was equal to that obtained with intrinsic factor-Cbl alone. The ability of EDTA to inhibit the binding of intrinsic factor-Cbl to membrane-bound (8) a n d solubilized preparations (4) of crude receptor has been shown previously.

Assay for Soluble
Purification of the Receptor-A s u m m a r y of a typical purification of receptor from 2.5 kg of canine ileal mucosa is shown i n FIG. 1. Assay for solubilized ileal receptor for intrinsic factor-Cbl. Assays were performed as described under "Methods" in which 1 pmol of hog intrinsic factor (IF)-["'Co]Cbl (specific activity, 0.5 to 1.0 pCi/pg of intrinsic factor) was incubated alone, M , or together, W, with 0.23 pmol of purified canine ileal receptor for intrinsic factor-Cbl. Various concentrations of (NH4)S04 were then added to determine the amount of hog intrinsic factor-["CoICbl precipitated under these conditions. The inset illustrates the values obtained for the differences between samples containing receptor and those not containing receptor a t each (NH.,)S04 concentration. hog intrinsic factor-Sepharose column without clogging the column. Most of the purification was obtained with the immunoaffinity chromatography procedure in which receptor saturated with hog intrinsic factor-Cbl was adsorbed to rabbit anti-hog intrinsic factor-Sepharose. The receptor was subsequently eluted with EDTA at pH 5.0 leaving behind the hog intrinsic factor-Cbl which remained attached to the anti-hog intrinsic factor-Sepharose. Preliminary experiments performed with hog intrinsic factor-Cbl covalently attached to Sepharose through either the intrinsic factor or the Cbl moieties resulted in the binding of much less receptor than was observed with the immunoaffinity chromatography technique that was finally adopted.
Gel Filtration of the Receptor-As shown in Fig. 2, two peaks of intrinsic factor-["CoICbl-binding ability were observed when partially purified (Fig. 2 A ) or purified (Fig. 2B) receptor were chromatographed on Bio-Gel A-5m in the presence of 0.15% Triton X-100. Based on the elution profiles of proteins of known molecular weight (see "Methods"), the two peaks of receptor had apparent molecular weights of 7,500,000 and 5,000,000 and indicated that the receptor had aggregated under the conditions utilized for the gel filtration. Similar elution profiles were observed for receptor-hog intrinsic factor-["CoICbl (data not presented) and for '2,"I-labeled receptor (Fig. 2C).
Polyacrylamide Disc Gel Electrophoresis-Evidence for receptor aggregation was also obtained with polyacrylamide disc gel electrophoresis in the presence of 0.1s% Triton X-100 since, as shown in Fig. 3, the receptor barely entered a 4% gel. A single protein band was observed approximately 2 mm from the top of the gel with diffuse staining occurring toward the cathode over an adjoining 6 mm. Both the major band and the diffuse area contained receptor activity as determined by measuring intrinsic factor-['CoICbl-binding ability in individually eluted gel slices (see Fig. 3 ) . The band and diffuse area did not stain for carbohydrate (data not shown) suggesting that the receptor has a very low or absent carbohydrate content. Attempts to obtain a sharper protein band by varying the time and voltage conditions of electrophoresis, by varying the concentration of Triton X-100, or by preincubating the receptor with Triton X-100 for 24 h prior to electrophoresis, proved ineffective.
SDS-Polyacrylamide Gel Electrophoresis-When Triton X-100 was removed from preparations of purified receptor and the receptor was subjected to polyacrylamide gel electrophoresis with 1% SDS, one protein band was observed as shown in Fig. 4 which had an apparent molecular weight of 180,000. Upon reduction with 1% 2-mercaptoethanol, the receptor gave two protein bands with apparent molecular weights of 59,000 and 42,000. These same bands should also be demonstrated when the receptor fraction was treated with a protease inhibitor before reduction and treatment with SDS (data not shown). The value of 180,000 is close to the value of 222,000 g of amino acid/mol of intrinsic factor-Cbl-binding ?'Ilabeled receptor (IO" dpm) was applied and individual gel slices were assayed for '"1. Middle, 35 pg of receptor protein was applied and individual gel slices were eluted and assayed for hog intrinsic factor-["CoICbl-binding ability. Although recoveries cannot be measured with precision, about 5-10% of the applied activity was eluted from the gel. Bottom, 35 pg of receptor protein was applied and the gel was stained for protein with Coomassie brilliant blue. In each case, the direction of electrophoresis was from left to right.

42,000
FIG. 4. SDS polyacrylamide gel electrophoresis of purified canine ileal receptor for intrinsic factor-Cbl performed in the absence, left, and in the presence, right, of 2-mercaptoethanol. Samples contain 25 pg of receptor protein and gels were stained for protein with Coomassie brilliant blue. The apparent molecular weights of the protein bands were determined with proteins of known molecular weight, as described under "Methods." The direction of electrophoresis was from top to bottom. ability obtained by amino acid analysis (see below) and supports the concept that the much higher molecular weights observed by gel filtration (see above) are due to receptor aggregation. The receptor appears to consist of two different subunits that are joined by disulfide bonds although the number of each subunit present is not known.
Isoelectric Focusing-When purified receptor was subjected to isoelectric focusing in the presence of 4 M urea and 2% Triton X-100, two protein-staining bands were observed with PI values of 4.9 and 5.4. These two bands could not be subdivided further when focusing was done for longer periods of time.
Amino Acid a n d Amino Sugar Composition-The results of the amino acid and amino sugar analyses are presented in Table 11. The purified receptor contained 222,000 g of amino acid and only 1,300 g of amino sugar/mol of intrinsic factor-Cbl-binding ability. The low value obtained for amino sugars suggests that the receptor contains at most a small amount of carbohydrate. The value of 222,000 g of amino acid/mol of intrinsic factor-Cbl-binding ability is close to the molecular weight of 180,000 obtained by SDS-polyacrylamide gel electrophoresis (see above) and indicates that the receptor contains a single intrinsic factor-Cbl binding site. The purified receptor contains approximately 57% of hydrophobic amino acids.
Affinity a n d Specificity of Purified Receptor-As shown in Fig. 5, purified canine receptor bound canine intrinsic factor-["Co]Cbl, hog intrinsic factor-["Co]Cbl, and human intrinsic factor-['CoICbl with association constants of 1.2 n"' , 3.2 n"' , and 4.0 n"' , respectively. These values are similar to those observed previously (8) ['CoICbl (19) was bound but with an affinity that was approximately 50-fold less than that observed with normal human intrinsic factor-['CoICbl, a difference that is similar to that observed with membrane bound human and canine acid/mol of intrinsic factor-Cbl-binding ability.

Ileal Receptor for
Intrinsic . The binding of hog intrinsic factor-["CoICbl was not inhibited by a 100-fold excess of nonradioactive Cbl nor by a 100-fold excess of hog intrinsic factor devoid of Cbl (data not presented). These studies indicate that binding to the purified canine receptor is specific for the intrinsic factor-Cbl complex as has been shown previously (8) for the crude membrane bound canine and human ileal receptors. Free Cbl or intrinsic factor or abnormal intrinsic factor-Cbl has very little or no affinity for purified receptor.
Studies with Anti-receptor Antibody-The data in Fig. 6 demonstrate first that ""I-labeled receptor was precipitated, with only 0.0625 p1 of rabbit anti-canine receptor antiserum. Approximately 75% of the 1251-labeled receptor was precipitated with 0.75 p1 of anti-receptor antiserum. In a separate experiment, when unlabeled receptor was used, hog intrinsic factor-["CoICbl-binding ability was also inhibited with only 0.0625 p1 of rabbit anti-canine receptor serum and 100% of the hog intrinsic factor-["Co]Cbl-binding ability was inhibited with 0.75 p1 of anti-receptor antiserum. Approximately 50% precipitation and inhibition were observed with 0.125 p1 of anti-receptor antiserum.
The anti-receptor antiserum contained a small amount of anti-intrinsic factor activity with 5 pl, 15 p1, and 75 pl, of antiserum giving 50% precipitation in 1.5 M (NHJSO, with canine intrinsic factor-["CoICbl, hog intrinsic factor-["Col-Cbl, and human intrinsic factor-["CoICbl, respectively. This anti-intrinsic factor activity may be due to the fact that the final preparation of purified receptor may have contained traces of intrinsic factor. This possibility is supported by the fact that the anti-receptor antiserum has more activity for canine and hog intrinsic factors than it does for human intrinsic factor, since canine intrinsic factor may have been present in the initial ileal homogenates while hog intrinsic factor was utilized directly in the purification of the receptor. It is also possible that the receptor for intrinsic factor-Cbl and intrinsic factor share at least one common antigenic determinant. In order to explore this latter possibility, we covalently Reactions of rabbit anti-canine ileal receptor for intrinsic factor-Cbl antiserum with purified receptor. '"'I-labeled receptor (25,000 dpm) and various amounts of antiserum were incubated in a volume of 500 pl for 48 h a t 4°C followed by centrifugation a t 20,000 X g for 30 min, and the percentage of '"I (0) precipitated was determined. In a separate experiment, ""I-receptor was replaced with 0.5 pmol of purified receptor and the supernatants were assayed for binding of hog intrinsic factor-["'CoICbl, (O), as described under "Methods." coupled rabbit anti-canine intrinsic factor antiserum, rabbit anti-hog intrinsic factor antiserum, rabbit anti-human intrinsic factor antiserum, and rabbit control serum to Sepharose. Individual columns containing 1 ml of each of these materials did not adsorb any purified receptor when 1 ml containing 5 pmol of purified receptor was passed over them in the presence or either 5 mM CaC12 or 5 mM EDTA, based on assays of the effluents for hog intrinsic factor-["CoICbl-binding ability. Control serum-Sepharose also failed to adsorb any of 5 pmol of canine, hog, or human intrinsic factor-["Co]Cbl while all three species of intrinsic factor-['CoICbl were adsorbed (>go%) by all three species of anti-intrinsic factor-Sepharose. These results suggest that the canine receptor for intrinsic factor-Cbl does not share antigenic determinants with canine, hog, or human intrinsic factor although they do not rule out this possibility.

DISCUSSION
Using immunoaffinity chromatography as the major purification technique, we have isolated the canine ileal receptor for intrinsic factor-Cbl in apparent homogeneous form. The final preparation bound 2.5 nmol of hog intrinsic factor-["'Co]Cbl/mg of protein and this represents a 10-fold and 25fold higher specific activity than previous purifications of the hog (15) and guinea pig (14) ileal receptors, respectively. The availability of milligram amounts of purified receptor has enabled us to perform both SDS-gel electrophoresis experiments and quantitative amino acid analyses. These experiments demonstrate that the receptor has a molecular weight of approximately 200,000, that it contains a single binding site for intrinsic factor-Cbl, that it contains little if any carbohydrate, and that the molecular weight values in excess of 1,000,000 obtained by gel fdtration are due to receptor aggregation. The purified receptor is essentially identical with preparations of unpurified membrane-bound receptor in terms of its affinity and specificity for intrinsic factor-Cbl (8) and in terms of its inhibition by EDTA (8). These properties indicate that the receptor has not been altered appreciably during any of the purification procedures.
Our studies indicate that reduction with 8-mercaptoethanol dissociates the canine ileal receptor for intrinsic factor-Cbl into two subunits with molecular weights of approximately 59,000 and 42,000 based on SDS electrophoresis. A previous study (15) has suggested that the hog ileal receptor for intrin-

Ileal
Receptor for Intrinsic Factor-Cobalamin sic factor-Cbl also contains two subunits, although the molecular weights obtained were 130,000 and 70,000. The differences observed between the canine and hog receptors may reflect species differences or could be due to the fact that Triton X-100 was not removed exhaustively from the hog receptor preparations (15) prior to SDS-gel electrophoresis in the presence of 2-mercaptoethanol. Failure to remove uncharged Triton X-100 molecules completely prior to electrophoresis could result in decreased binding of negatively charged SDS molecules with resultant decreased protein mobilities during electrophoresis and a resultant overestimation of molecular weight. These differences could also be due to the fact that the specific activity of the purified canine ileal receptor (see "Results") is approximately 10-fold higher than that of the purified hog ileal receptor (15) which only bound 0.25 nmol of hog intrinsic factor-["Co]Cbl/mg of protein. Thus, the 130,000 and 70,000 molecular weight proteins observed with the hog ileal receptor may actually represent contaminants rather than the receptor itself. This possibility, or species differences, may also explain why the hog ileal receptor (15) appeared to stain for carbohydrate while the canine ileal receptor did not stain for carbohydrate and contained less than 1% amino sugar.
Previous studies (21,27) have shown that intrinsic factor, R protein, transcobalamin 11, and the transcobalamin I1 receptor, resemble one another in terms of having single polypeptide chains with molecular weights of approximately 40,000 for their amino acid portions, single binding sites, and similarities in amino acid composition. These similarities have suggested that proteins involved in Cbl transport may have evolved from a common ancestral gene, although differences in amino acid composition, isolated congenital deficiencies of these proteins, and an apparent lack of immunologic cross-reactivity demonstrate that each is now coded for by a separate gene (27). The canine ileal receptor for intrinsic factor-Cbl is clearly different from intrinsic factor since the receptor has a molecular weight of approximately 200,000, contains two polypeptide chains that appear to be joined via disulfide bonds, and differs markedly from intrinsic factor in its content of several amino acids and amino sugars. One of the subunits does, however, have a molecular weight of 42,000 and it would be interesting to study this subunit in more detail to determine its amino acid composition and whether this subunit possesses the binding site for intrinsic factor-Cbl.
We have prepared anti-canine receptor antibodies in rabbits and have shown that they both precipitate the canine ileal receptor and inhibit its binding of intrinsic factor-Cbl. A small amount of activity is observed against intrinsic factor from various species, but it appears likely (see "Results") that small amounts of anti-intrinsic factor antibody were raised against trace amounts of intrinsic factor that may have been present in the purified receptor preparations, since three different antisera raised against canine, hog, and human intrinsic factor did not react with the canine ileal receptor. A previous study (15) has suggested that anti-intrinsic factor antibodies may cross-react with a t least one component present in purified hog ileal receptor preparations, but details concerning these experiments were not provided and the possibility that the component recognized was actually intrinsic factor itself was not excluded. In any event, it appears likely that partially purified or adsorbed preparations of anti-receptor antibodies will be useful in future studies designed to elucidate the subcellular localization of the receptor for intrinsic factor-Cbl and the role that the receptor plays in the cellular uptake of Cbl by the ileum and its subsequent release into plasma.