Isolation and characterization of an abnormal human intrinsic factor

The Neurology of Cobalamin

The following review indicates that the impact of cobalamin on neurologic disease extends far beyond the traditional myelopathy of classical pernicious anemia. The delineation of a broad spectrum of inherited disorders of cobalamin processing has served to illustrate and precisely define each step in the normal absorption, transport and intracellular metabolism of this essential vitamin. Recent clinical work has extended the boundaries of acquired cobalamin deficiency to encompass a variety of neuropsychiatric disturbances without identifiable concomitant hematologic derangements and emphasized the utility and sensitivity of new laboratory tests. These findings will demand increased vigilance from clinicians so that atypical and subtle cobalamin deficiency states will be readily diagnosed. The wide range of neurologic dysfunction observed in both inherited and acquired disorders of cobalamin metabolism challenges basic scientists to delineate cobalamin’s presumed important role in the normal development and homeostasis of the nervous system.

Diagnosis of megaloblastic anaemias

There are a large number of causes of megaloblastic anaemia. The most frequent are disorders resulting in vitamin B(12) or folate deficiency. The diagnostic process often consists first of establishing the presence of B(12) or folate deficiency and then of determining the cause of deficiency. The blood count, blood film, serum B(12) assay, and red cell and serum folate assays are the primary investigations. Other useful investigations include serum/plasma methylmalonic acid (MMA), plasma total homocysteine (tHCYS) and serum holo-transcobalamin II assays. All currently used tests have limitations regarding specificity or sensitivity or both and the metabolite assays are not widely available. An understanding of these limitations is essential in formulating any diagnostic strategy. The wide use of serum B(12) and metabolite assays has resulted in the increasingly early diagnosis of B(12) deficiency, often in patients without B(12)-related symptoms (subclinical deficiency). Food cobalamin malabsorption is the most frequent cause of a low serum B(12). At least 25% of low serum B(12) levels are not associated with elevated metabolite levels and may not indicate B(12) deficiency. Some of these are caused by partial deficiency of transcobalamine I.

Hereditary juvenile cobalamin deficiency caused by mutations in the intrinsic factor gene

Hereditary juvenile megaloblastic anemia due to vitamin B12 (cobalamin) deficiency is caused by intestinal malabsorption of cobalamin. In Imerslund-Grasbeck syndrome (IGS), cobalamin absorption is completely abolished and not corrected by the administration of intrinsic factor (IF); if untreated, the disease is fatal. Biallelic mutations either in the cubilin (CUBN) or amnionless (AMN) gene cause IGS. In a series of families clinically diagnosed with likely IGS, at least six displayed no evidence of mutations in CUBN or AMN. A genome-wide search for linkage followed by mutational analysis of candidate genes was performed in five of these families. A region in chromosome 11 showed evidence of linkage in four families. The gastric IF (GIF) gene located in this region harbored homozygous nonsense and missense mutations in these four families and in three additional families. The disease in these cases therefore should be classified as hereditary IF deficiency. Clinically, these patients resembled those with typical IGS; radiocobalamin absorption tests had been inconclusive regarding the nature of the defect. In the diagnosis of juvenile cobalamin deficiency, mutational analysis of the CUBN, AMN, and GIF genes provides a molecular characterization of the underlying defect and may be the diagnostic method of choice.

A genetic polymorphism in the coding region of the gastric intrinsic factor gene (GIF) is associated with congenital intrinsic factor deficiency

Congenital intrinsic factor (IF) deficiency is a disorder characterized by megaloblastic anemia due to the absence of gastric IF (GIF, GenBank NM_005142) and GIF antibodies, with probable autosomal recessive inheritance. Most of the reported patients are isolated cases without genetic studies of the parents or siblings. Complete exonic sequences were determined from the PCR products generated from genomic DNA of five affected individuals. All probands had the identical variant (g.68A>G) in the second position of the fifth codon in the coding sequence of the gene that introduces a restriction enzyme site for Msp I and predicts a change in the mature protein from glutamine(5) (CAG) to arginine(5) (CGG). Three subjects were homozygous for this base exchange and two subjects were heterozygous, one of which was apparently a compound heterozygote at positions 1 and 2 of the fifth codon ([g.67C>G] + [g.68A>G]). The other patient, heterozygous for position 2, had one heterozygous unaffected parent. Most parents were heterozygous for this base exchange, confirming the pattern of autosomal recessive inheritance for congenital IF deficiency. cDNA encoding GIF was mutated at base pair g.68 (A>G) and expressed in COS-7 cells. The apparent size, secretion rate, and sensitivity to pepsin hydrolysis of the expressed IF were similar to native IF. The allelic frequency of g.68A>G was 0.067 and 0.038 in two control populations. This sequence aberration is not the cause of the phenotype, but is associated with the genotype of congenital IF deficiency and could serve as a marker for inheritance of this disorder.

Sequence, S-S bridges, and spectra of bovine transcobalamin expressed in Pichia pastoris

Transcobalamin (TC) -encoding cDNA was isolated from a bovine mammary gland cDNA library. Hybridization of the cloned bovine TC-cDNA to RNA samples from bovine tissues showed that the most intensive synthesis of a TC positive 1.9-kilobase mRNA occurred in kidney, lymphatic nodes, and liver. Bovine TC was expressed in yeast Pichia pastoris, and the isolated recombinant protein showed cobalamin (Cbl) and receptor binding properties similar to TCs from other sources. Alignment of the related Cbl carriers (haptocorrins and intrinsic factors from other species) with bovine TC (414 residues) revealed four conservative clusters in the sequence (85-98, 137-147, 178-190, and 268-288), which may be responsible for Cbl binding. Three S-S bonds connected Cys residues 3-252, 98-294, and 147-190. Treatment with an S-S reducing agent caused liberation of Cbl from TC-Cbl. A significant change was observed in the TC-Cbl absorbance spectrum upon substitution of Co(2+)-coordinated H(2)O by azide. The reaction developed several orders of magnitude slower, and the spectral distortions were much stronger than those in free Cbl. This may be caused by significant deformation of the Cbl molecule and/or by its shielding when bound to TC.

Megalin-mediated endocytosis of transcobalamin-vitamin-B12 complexes suggests a role of the receptor in vitamin-B12 homeostasis

Kidney cortex is a main target for circulating vitamin B12 (cobalamin) in complex with transcobalamin (TC). Ligand blotting of rabbit kidney cortex with rabbit 125I-TC-B12 and human TC-57Co-B12 revealed an exclusive binding to megalin, a 600-kDa endocytic receptor present in renal proximal tubule epithelium and other absorptive epithelia. The binding was Ca2+ dependent and inhibited by receptor-associated protein (RAP). Surface plasmon resonance analysis demonstrated a high-affinity interaction between purified rabbit megalin and rabbit TC-B12 but no measurable affinity of the vitamin complex for the homologous alpha 2-macroglobulin receptor (alpha 2MR)/low density lipoprotein receptor related protein (LRP). 125I-TC-B12 was efficiently endocytosed in a RAP-inhibitable manner in megalin-expressing rat yolk sac carcinoma cells and in vivo microperfused rat proximal tubules. The radioactivity in the tubules localized to the endocytic compartments and a similar apical distribution in the proximal tubules was demonstrated after intravenous injection of 125I-TC-B12. The TC-B12 binding sites in the proximal tubule epithelium colocalized with megalin as shown by ligand binding to cryosections of rat kidney cortex, and the binding was inhibited by anti-megalin polyclonal antibody, EDTA, and RAP. These data show a novel nutritional dimension of megalin as a receptor involved in the cellular uptake of vitamin B12. The expression of megalin in absorptive epithelia in the kidney and other tissues including yolk sac and placenta suggests a role of the receptor in vitamin B12 homeostasis and fetal vitamin B12 supply.

Multiplex reverse transcription polymerase chain reaction combined with temperature gradient gel electrophoresis as a tool for the normalized quantitation of intrinsic factor mRNA

For the quantitation of intrinsic factor (IF) mRNA, an assay based on competitive reverse transcription and subsequent polymerase chain reaction (RT-PCR) combined with temperature gradient gel electrophoresis (TGGE) was established and validated with respect to precision and accuracy. IF-specific mRNA segments ("targets") were coamplified with known amounts of homologous "standard" RNA molecules, which differed from the targets by one base substitution. Following amplification, TGGE heteroduplex analysis proved to be a powerful method facilitating the efficient separation of these nearly identical target and standard DNA products. The measured absolute copy numbers of IF mRNA were put into relation to the constitutively expressed mRNA specific for glyceraldehyde-3-phosphate dehydrogenase (GAPDH), quantified simultaneously by competitive multiplex RT-PCR. The resulting normalized IF mRNA expression rate in terms of n copies of IF mRNA/copy of GAPDH mRNA is independent of the mRNA heterogeneity and the abundance of specific transcripts within the RNA population of interest. Therefore, normalization relative to the housekeeping gene GAPDH provides a widely applicable value for comparative studies of gene expression on the level of mRNA. Here, a normalized IF mRNA expression rate of three copies per GAPDH mRNA molecule was measured in human stomach mucosa.

Cobalamin malabsorption in three siblings due to an abnormal intrinsic factor that is markedly susceptible to acid and proteolysis

Three siblings presented in their second year of life with megaloblastic anemia that responded to parenteral cobalamin (Cbl). Schilling tests were less than 1%, correcting to 5 to 15% after addition of hog intrinsic factor (IF). Gastric acid analysis and gastric biopsies were normal by light and electron microscopy. Gastric juice contained less than 3 pmol/ml of Cbl-binding ability due to IF (normal, 10-34 pmol/ml) and less than 2 pmol/ml of IF when measured with a radioimmunoassay (RIA) using normal human IF-[57Co]Cbl and rabbit anti-human IF serum (normal, 17-66 pmol/ml). However, RIA employing rabbit anti-hog IF serum gave values of 4-13 pmol/ml of IF (normal, 11-33 pmol/ml). This material had an apparent molecular weight of 40,000 (normal IF = 70,000). The IF from gastric biopsies appeared normal in terms of Cbl-binding ability, ileal binding, molecular weight, and both RIAs. This IF differed from normal mucosal IF, in that it lost its Cbl-binding ability when incubated at 37 degrees C at acid pH or in the presence of pepsin or trypsin. This loss was retarded when [57Co]Cbl was bound to the IF before these incubations. The stabilizing effects of neutralization and Cbl were also demonstrated in vivo. Schilling tests for the siblings of 0.4, 0.5, and 1.0% increased to 2.7, 5.7, and 4.3% (P less than 0.05), respectively, when the Schilling tests were repeated with the addition of NaHCO3 and cobinamide (which allows Cbl to bind immediately to IF). We conclude that Cbl malabsorption in these children is due to an abnormal IF that is markedly susceptible to acid and proteolytic enzymes which cause a decrease in its molecular weight and Cbl-binding ability and a loss of antigenic determinants that are recognized by the anti-human IF serum.

Intrinsic factor within parietal cells of patients with juvenile pernicious anemia. A retrospective immunohistochemical study

One of the diverse group of disorders that cause pernicious anemia in childhood, juvenile pernicious anemia, has been characterized by normal acid secretion, normal gastric mucosal histology, adequate intestinal absorption of cobalamin in the presence of exogenous gastric intrinsic factor (IF), but the inadequate "production" of IF from birth. Inadequate production has been inferred from the absence of measured IF in stimulated gastric secretions. To assess whether immunogenic IF was commonly present within the parietal cells of subjects with juvenile pernicious anemia, we studied paraffin-embedded biopsy material from the largest reported series of childhood pernicious anemia, using a well-characterized indirect immunoperoxidase method. Preliminary studies were able to identify IF in fundic mucosal biopsy specimens that had been stored for as long as 27 yr. In a blinded evaluation, six of the nine fundic biopsy specimens from children with juvenile pernicious anemia demonstrated immunogenic IF. Two sets of siblings were concordant for the presence or absence of intracellular IF, and six gastric biopsy specimens from patients with Imerslund's syndrome were all positive for IF. These findings indicate that juvenile pernicious anemia is a heterogeneous group of disorders whose similar clinical expression might be caused by (a) inadequate synthesis of IF, (b) a block in IF secretion, (c) the secretion of an abnormal IF that does not bind to cobalamin, or (d) the secretion of other abnormal IFs that could contain a number of other functional defects.

Gastric juice in congenital pernicious anemia contains no immunoreactive intrinsic factor molecule: study of three kindreds with variable ages at presentation, including a patient first diagnosed in adulthood

The mechanism responsible for the isolated intrinsic factor deficiency in congenital pernicious anemia is unknown. A new second-antibody radioimmunoassay capable of recognizing intrinsic factor independent of the molecule's ability to bind added cobalamin was used to study six patients from three kindreds with this disorder. One of the patients was first diagnosed at age 23 because of unusual circumstances in her case; yet the other patients also demonstrated great age variability at presentation of this presumably congenital disorder, even within the same kindred. The radioimmunoassay failed to detect immunoreactive intrinsic factor in any of the six patients, suggesting that elaboration of an abnormal molecule was not the pathogenetic mechanism. An unexpected incidental finding, contrasting with this observation in congenital pernicious anemia, was immunologic evidence that a previously described patient with familial R binder deficiency clearly elaborated an abnormal R binder molecule.

Human intrinsic factor secretion: immunocytochemical demonstration of membrane-associated vesicular transport in parietal cells

The human gastric parietal cell synthesizes and secretes intrinsic factor (IF) and acid. In contrast to the cellular mechanisms of acid secretion, little is known about the mechanisms of IF secretion. To elucidate these mechanisms we obtained gastric secretions and sequential fundic biopsies from three subjects before and after pentagastrin stimulation (6 microgram/Kg s.c.). IF was localized in the biopsies using an ultrastructural immunoperoxidase technique using a well-characterized, monospecific antibody to human IF. IF output was quantified using a specific radioimmunoassay in concurrently obtained gastric secretions. Before stimulation, IF was associated with tubulovesicles scattered throughout the cytoplasm and with some in rough endoplasmic reticulum (RER). The tubulovesicles associated with IF migrated to the periphery of the secretory canaliculi within 8 min of stimulation. IF was present on secretory microvilli between 8 and 30 min when IF output in gastric juice was at its maximum. The cessation of IF secretion coincided with the depletion of IF associated with tubulovesicles. IF appeared in the perinuclear space and RER as the IF associated with tubulovesicles was secreted. These observations indicate that IF secretion depends upon membrane-associated vesicular transport and provides support for a membrane translocation-fusion hypothesis to explain the morphologic changes that occur in the parietal cell during secretion.

Megaloblastic anemia as a result of an abnormal transcobalamin II (Cardeza)

A 34-year-old Black woman had severe megaloblastic anemia in childhood. Initially, and over the years, she responded well to massive doses of parenteral cobalamin (Cbl) or oral folic acid. Metabolic reactions involving Cbl and folate enzymes were normal during both relapse and remission except for the absence of thymidylate synthetase in relapse. Amino acid analyses of urine and plasma showed no significant abnormalities. Neither cystathionine, homocystine, formiminoglutamic acid, nor methylmalonic acid was detected in the urine. The serum Cbl level was repeatedly elevated even when the patient was receiving only folic acid therapy. The elevation of the vitamin in the serum was found to be a result of markedly increased levels of transcobalamin II (TC II), as identified by several physicochemical techniques. The patient's TC II-Cbl shared immunologic properties with normal TC II but did not facilitate or impede the uptake of Cbl or Cbl bound to normal TC II, respectively, by human cells.

Primary structure of human intrinsic factor: progress report on cyanogen bromide fragmentation

Human intrinsic factor purified by labile ligand affinity chromatography was cleaved with cyanogen bromide and fractionated by gel filtration. Four of the fragments were purified and sequenced to a total of eighty-four amino acid residues. Including the N-terminal amino acids this amounts to one third of the total amino acid sequence of human intrinsic factor. One of the fragments contained a tyrosine labelled only on iodination of intrinsic factor devoid of cobalamin.

Correction of cobalamin malabsorption in pancreatic insufficiency with a cobalamin analogue that binds with high affinity to R protein but not to intrinsic factor. In vivo evidence that a failure to partially degrade R protein is responsible for cobalamin malabsorption in pancreatic insufficiency

In vitro studies indicate that [(57)Co]cobalamin (Cbl) is preferentially bound to salivary R protein as opposed to intrinsic factor (IF) and that [(57)Co]Cbl bound to R protein is not transferred to IF at either pH 2 or pH 8. Incubation of R protein-[(57)Co]Cbl with pancreatic proteases causes a partial degradation of the R protein moiety and a rapid transfer of [(57)Co]Cbl to IF. We have postulated that the etiology of Cbl malabsorption in pancreatic insufficiency is an inability to partially degrade R protein because of a lack of pancreatic proteases. We have tested this hypothesis by determining the ability of a nonradioactive Cbl analogue, bound with high affinity by R protein but not by IF, to correct the malabsorption of [(57)Co]Cbl in patients with pancreatic insufficiency.R protein bound the Cbl analogue known as cobinamide with affinities that were the same and only 14-fold lower than those for Cbl at pH 8 and pH 2, respectively. Cobinamide was bound by IF with affinities that were 600,000- and 10,000-fold lower than those for Cbl at pH 8 and 2, respectively. The addition of 125 pmol of nonradioactive cobinamide to 0.5 pmol of [(57)Co]Cbl before being added to 1 pmol of R protein and 1 pmol of IF, markedly inhibited the ability of R protein to compete with IF for binding the [(57)Co]Cbl. Similar results were obtained with freshly aspirated gastric juice. This change was essentially indistinguishable from that observed previously when R protein or R protein-[(57)Co]Cbl was incubated in vitro with trypsin. The oral administration of 100 nmol of nonradioactive cobinamide in Schilling tests was equivalent to trypsin in its ability to completely correct the malabsorption of 0.4 nmol of [(57)Co]Cbl in three patients with pancreatic insufficiency. The fact that both trypsin and nonradioactive cobinamide inhibit the ability of R protein to compete with IF for [(57)Co]Cbl binding in vitro, and correct the mal-absorption of [(57)Co]Cbl in patients with pancreatic insufficiency in vivo, supports our hypothesis that the primary defect in Cbl absorption in this disease is an inability to partially degrade R protein because of a lack of pancreatic proteases.

Effect of proteolytic enzymes on the binding of cobalamin to R protein and intrinsic factor. In vitro evidence that a failure to partially degrade R protein is responsible for cobalamin malabsorption in pancreatic insufficiency

Cobalamin (Cbl; vitamin B(12)) malabsorption in pancreatic insufficiency can be partially corrected by bicarbonate and completely corrected by pancreatic proteases but the mechanisms involved are unknown. Because saliva contains enough R-type Cbl-binding protein (R protein) to bind all of the dietary and biliary Cbl, it is possible that R protein acts as an inhibitor of Cbl absorption and that pancreatic proteases are required to alter R protein and prevent such inhibition. To test this hypothesis we studied the ability of R protein and intrinsic factor (IF) to compete for Cbl binding and ability of pancreatic proteases to alter this competition. Human salivary R protein bound Cbl with affinities that were 50- and 3-fold higher than those of human IF at pH 2 and 8, respectively. Cbl bound to IF was transferred to an equal amount of R protein with t((1/2))'s of 2 and 90 min at pH 2 and 8, respectively, and within several hours respective ratios of R protein-Cbl/IF-Cbl of 50 and 2 were observed. Cbl bound to R protein was not transferred to IF at either pH 2 or 8. Incubation of R protein with pancreatic proteases at pH 8 led to a 150-fold decrease in its affinity for Cbl. Incubation of R protein-Cbl with pancreatic proteases led to complete transfer of Cbl to IF within 10 min. Gel filtration studies with R protein-[(57)Co]Cbl and (125)I-R protein showed that pancreatic proteases partially degraded R protein. Pancreatic proteases differed in their ability to effect these changes with trypsin > chymotrypsin > elastase. Pancreatic proteases did not alter IF in any of the parameters mentioned above. Pepsin failed to alter either R protein or IF. THESE STUDIES SUGGEST THE FOLLOWING: (a) that Cbl is bound almost exclusively to R protein in the acid milieu of the stomach, rather than to IF as has been assumed previously; (b) that Cbl remains bound to R protein in the slightly alkaline environment of the intestine until pancreatic proteases partially degrade R protein and enable Cbl to become bound exclusively to IF; and (c) that the primary defect in Cbl absorption in pancreatic insufficiency is a lack of pancreatic proteases and a failure to alter R protein and effect the transfer of Cbl to IF. These studies also suggest that the partial correction of Cbl malabsorption observed with bicarbonate is due to neutralization of gastric HCl, since at slightly alkaline, pH IF can partially compete with R protein for the initial binding and retention of Cbl.

Isolation and characterization of a novel vitamin B12-binding protein associated with hepatocellular carcinoma

High levels of a novel vitamin B12-binding protein (hepatoma B12 BP) have been observed recently in plasma obtained from three adolescent patients with hepatocellular carcinoma. This protein has now been isolated in homogeneous form from the plasma and pleural fluid of two of these patients by the use of affinity chromatography with vitamin B12-Sepharose. The hepatoma B12 BP belongs to the R-type group of B12-binding proteins and is essentially indistinguishable from the recently isolated human milk and saliva R-type proteins in terms of: (a) immunologic properties based on immunodiffusion and immunoprecipitation assays; (b) amino acid composition; (c) molecular weight based on amino acid and carbohydrate content; and (d) absorption spectra. Both hepatoma B12 BPs contain more sialic acid and less fucose than the milk and saliva B12 BPs. All four proteins contain similar amounts of galactose, mannose, galactosamine, and glucosamine. Differences in sialic acid content appear to account for the differences in electrophoretic mobility that were observed among the four proteins. Differences in total carbohydrate content appear to account for the differences in apparent molecular weight that were observed with both gel filtration and sodium dodecyl sulfate-polyacrylamide gel electrophoresis. Tumor tissue from one of the patients contained 10 times as much R-type protein as did normal liver tissue from the same patient. This suggests, although it does not prove, that synthesis by the tumor is the cause of the high levels of R-type protein found in the plasma of certain patients with hepatocellular carcinoma. Plasma survival studies performed with rabbits indicate that the hepatoma B12 BP has a prolonged plasma survival and suggests that his parameter is also of importance.