Isolation, Purification, and Characterization of an Amadori Product Binding Protein from a Pseudomonas sp. Soil Strain”

Sugars react nonenzymatically with protein amino groups to form a ketoamine adduct (Amadori product), which leads to the formation of advanced glycation end- products. These compounds are involved in the develop-ment of tissue modifications such as cross-linking and fluorescence in diabetes and aging. Searching for an enzyme to reverse protein glycation, we isolated a Pseudo- monm sp. soil strain growing selectively on the Amadori product e-fructosyl-aminocaproate. An Amadori product binding protein (ABP) was purified from the bacterial extract by single-step affinity chromatography on gly- cated lysine-Sepharose. The protein, a monomer of 45 kDa, did not bind to unglycated or NaBHpeduced gly- cated lysine-Sepharose suggesting specificity for the Amadori compound. The concentration-dependent binding of glycated aminocaproate showed saturation with Kd = 1.49 p~ and B,, = 17.63 nmoYmg of protein corre- sponding to 0.8 moYmol of protein. The binding of e-l-[14C]fructosyl-aminocaproate to the protein was inhib- ited by other glucose-derived Amadori products, but not by free sugars, unglycated amines, or ribated lysine. The sequence of the first 16 NH,-terminal amino acids and a GenBank search revealed that ABP is a novel protein. Its synthesis was inducible the of an excess of unlabeled fructosyl-e-aminocaproate (200 After incubation at 4 "C for 20-22 h in a rotary shaker, two aliquots of 20 pl were withdrawn from each well, and the radioactivity was measured. For the inhibition studies, the '"C-labeled Amadori product was used at 6 PM constant concentration with and without a 10-fold molar concentration of the compound to be tested. The ability ofABP to bind different compounds was expressed as percent of the binding of e-[U-"Clfructosyl-aminocaproate. For the saturation studied, the "C-labeled Amadori product was used at concentration ranging from 0.5 to 18 p . A plot of bound Amadori concentration was obtained, and the binding parameters were calculated according to the Michaelis- Menten equation using nonlinear fitting analysis. (ELISA) for Binding to Glycated Proteins-For investigation of the ability of ABP to bind to glycated proteins, an ELISA sandwich assay was performed. Briefly, glycated proteins (BSA and were immobilized on microtiter plates, and after blocking with unglycated BSA, ABP (10-100 ng) was added to each well. Following 16 h of incubation at 4 "C, bound ABP was detected by ELISA using a monoclonal antibody ABP (see below) and goat

In recent years, pharmacological approaches for blocking the advanced Maillard reaction have been developed. Aminoguanidine, an inhibitor of AGE formation (12, 131, prevented the increase in regional blood flow (14) and basement membrane thickness (15), and ameliorated retinopathy (16) and neuropathy (17) in diabetic rats.
The possibility of breaking off the chain of events resulting from nonenzymatic glycation by enzymatically reversing the Amadori product could be a powerful tool to fully understand the role of nonenzymatic glycation in diabetes and aging. Searching for such approaches as a potential gene therapy of complications resulting from glycation, we have grown soil organisms on glycated substrate as sole carbon source hoping to isolate enzyme(s) that degrade Amadori products. We describe here the isolation and characterization of a novel protein from a Pseudomonas soil sp. that binds selectively glycated amino acids.
(pH 3.25), and the effluent was monitored by thin layer chromatography on silica plates with 1-butanollacetic acidwater (40:6:10). Plates were developed both with ninhydrin and triphenyltetrazolium chloride. The fractions containing the Amadori product were pooled and flash chromatographed on a silica gel column (5 x 30 cm) with methanoVethy1 acetate (3:l). The eluted Amadori product was concentrated by rotary evaporation, redissolved with 150 ml of methanol and precipitated by dropwise addition into 2 liters of cold ethyl acetate. The purity was checked by thin layer chromatography and confirmed by proton NMR.
N"-t-Boc-W-1-deoxyfructosyl-lysine and W-acetyl-N"-1-deoxyfructosyl-lysine were synthesized as described by Finot and Mauron (18) and as above. el-deoxy-ribosyl lysine was synthesized as described previously (19). I-deoxy-fructosylglycine and 1-deoxy-fructosylpropylamine were a generous gift from Dr. T. Horiuchi and from Dr. Marcus Glomb, respectively. Borohhydride-reduced E-I-deoxyfructosyl aminocaproic acid was prepared through reduction in presence of 50 M excess NaBH,. The product was dissolved over a Sephadex G-10 column. Purity was checked by thin layer chromatography. A faint spot of original material was present beside the major spot of reduced material.
The Amadori product was eluted from the paper with water and lyophilized. The purity was assessed by thin layer chromatography followed by a radioactivity scanner from Berthold (Bad Wildbad, Germany).
Glycated BSA and glycated RNase were prepared by incubation of protein (50 mg/ml) with glucose in phosphate-buffered saline up to 30 days. After dialysis, the extent of glycation was evaluated by furosine measurement (20).

Microorganisms
The isolation of Amadori product utilizing microorganisms from soil samples, was performed by direct selection using e-I-deoxyfructosyl aminocaproate as sole carbon source. The culture medium used was a minimal salt (KH,P04 0.3 g/100 ml, GHPO, 0.7 g/lOO ml, (Na,),SO, 0.01 4100 ml, MgSO, 1 mM) containing biotin and thiamin (0.01 pg/ml), riboflavin (2.5 pg/ml), folic acid, pyridoxine, and pantothenic acid 0.5 pg/ml (minimal medium). Solid media containing 2 g of agar/100 ml of medium were also prepared. The Amadori product was added to a final concentration of 0.5 dl00 ml. As control, the same medium was supplemented with glucose or €-aminocaproate at the same molar concentration as the Amadori compound. Microorganisms were grown aerobically at 37 "C. Strains of various other bacterial species were obtained from the laboratory of clinical microbiology at the Institute of Pathology. All strains were maintained both at -80 "C in 32.5% glycerol and at 4 "C on solid medium with subcultures every 3 months.
Synthesis of Affinity Chromatography Support (Glycated Lysine-Sepharose) Glycated lysine-Sepharose was obtained by incubating lysine-Sepharose (4 pmol of NHJml gel) with 1 M glucose in 0.25 M sodium phosphate buffer (pH 7.4) at 50 "C for 64 h. After centrifugation, 0.5 M acetic acid was added to the packed gel in order to destroy the labile Schiff base. The gel was washed extensively with distilled water to remove free glucose and stored at 4 "C in 0.05 M sodium phosphate buffer (pH 7.4) until use. In some experiments 1 mM EDTA was added to the incubation mixture. As control, lysine-Sepharose was incubated under identical conditions except that glucose was omitted.
Deoxyglucitolyllysine-Sepharose was obtained by NaBH, reduction of glycated lysine-Sepharose. To 1 ml of 0.05 M sodium phosphate buffer was added a 20-fold molar excess of NaBH, from a 2 M stock solution in M NaOH. After 10 min at room temperature followed by 50 min on ice, the reaction was terminated by the addition of 100 pl of 6 N HCl. The gel was washed extensively with distilled water and stored in 0.05 M sodium phosphate buffer (pH 7.4) at 4 "C until use.
Chromatography of Cell Extract on Glycated Lysine-Sepharose All steps were performed at 4 "C. For the isolation of proteins binding to glycated lysine-Sepharose, cells from 100 ml of a 24-h culture were harvested by centrifugation a t 5,000 x g, washed once with minimal salt, resuspended with 1 ml of 0.01 M sodium phosphate buffer (pH 7.4) containing 1 mM EDTA and 0.2 nm phenylmethylsulfonyl fluoride, and disrupted by abrasion with glass beads. After centrifugation at 20,000 x g for 25 min, the clear supernatant was divided into two aliquots of 1 mg total protein each and applied to both a glycated and unglycated lysine-Sepharose column (1 ml gel) equilibrated with 0.05 M sodium phosphate buffer (pH 7.4). The column was washed with equilibrating buffer until the absorbance at 280 nm was <0.01 unit, followed by 0.2 M NaCl in the same buffer, and then eluted with 0.2 M fructosylaminocaproate. The eluted fractions were dialyzed twice against distilled water, lyophilized, resuspended with 50 p1 of water and analyzed by SDS-PAGE. As control, the same experiments were performed with extract from Escherichia coli (strain ATCC 15922) and Pseudomonas aeruginosa non mucoid strain (ATCC 27853) and with the soil strain cultivated in a complete medium (Nutrient Broth, Difco) or minimal medium supplemented with glucose and e-aminocaproic acid.
For preparative purposes, washed cells from 400 ml of culture were disrupted by three passes through a French press at 6,000-7,000 p.s.i., followed by sonication. Following centrifugation, the cell-free extract (10-15 mg of total proteins) was applied to a column of glycated lysine-Sepharose (6 ml of gel). After washing as above, the binding protein was eluted with 1 M NaCl in 0.05 M sodium phosphate buffer (pH 7.4) and concentrated to approximately 2 ml by ultrafiltration (Diaflo membrane YMIO) under nitrogen. The purity was verified by SDS-PAGE both with Coomassie Blue and silver staining detection.

Induction Experiment
To investigate the effect of the presence of Amadori product in the culture medium on the expression of proteids) binding to glycated lysine-Sepharose, cells harvested from 400 ml of a nutrient broth culture were washed once with minimal salt, resuspended with 150 ml of minimal medium supplemented with 0.5 g/lOO ml fructosylaminocaproate, and incubated a t 37 "C. At various time points, 20 ml of the cell suspension were withdrawn, washed with minimal salt, and resuspended with 0.8 ml of 0.01 M sodium phosphate buffer (pH 7.4) containing 1 mM EDTA and 0.2 mM phenylmethylsulfonyl fluoride. The cells were disrupted as above and centrifuged. An amount of each extract corresponding to 0.7 mg of total protein was chromatographed over a 1-ml column of glycated lysine-Sepharose as above, and the eluates were dialyzed, lyophilized, and analyzed by SDS-PAGE.

Binding Studies
The binding activity of the purified protein toward various substrates was characterized using three types of assays.
Binding to Glycated Lysine-Sepharose-In order to evaluate the relationship between extent of glycation of lysine-Sepharose and binding of the purified protein, a mixture of lysine-Sepharose (4 pmol of NHdml of gel) and glucose (1 M) was prepared as above and incubated at 50 "C. 1-ml aliquots were withdrawn at time intervals and washed to remove free glucose. A 0.5-ml gel column was packed with each glycated lysine-Sepharose preparation. The glucose incorporation was evaluated by incubation of 1 ml of the original mixture in the presence of [U-'4Clglucose (final specific activity, 1 mCi/mmol). At the same time points 100 pl were withdrawn and washed, and 25 pl of packed gel were counted in duplicate using a Beckman LS 6000Sc p-counter. 8 pg of purified protein were applied to each column preequilibrated with 0.05 M sodium phosphate buffer (pH 7.4). After washing with the same buffer, columns were eluted with 2 ml of 0.2 M fructosylaminocaproate. The eluates were concentrated by ultrafiltration (Centricon 10, Amicon, Inc.), and comparable volumes of each samples were subjected to SDS-PAGE. Following Coomassie Blue staining, the gel was analyzed by scanning densitometry (UBS SciScan 5000). The optical density of the bands and the corresponding value of incorporated radioactivity were plotted uersus time.
To test the binding specificity of the purified protein for the Amadori product, a series of 0.5-ml gel columns was prepared with lysine-Sepharose, Sepharose-fructosyllysine, Sepharose-fructosyllysine-EDTA and Sepharose-deoxyglucitollysine. Purified protein (8 pg) was applied to each column and chromatographed as above. Both the washes and the eluates were concentrated by ultrafiltration and analyzed by SDS-PAGE. Equilibrium Dialysis-An equilibrium dialysis radioligand binding assay was used to further characterize the binding specificity of ABP (inhibition studies) and to evaluate the binding parameters with the model Amadori product e-fmctosyl-aminocaproate (saturation studies). The assay was carried out in polypropylene microdialysis cells (Bel-Art Products, Pequannock, NJ) consisting of two wells (0.1 ml each) separated by a dialysis membrane disc. One well contained 10 pg of ABP in 60 pl of 0.05 M sodium phosphate buffer (pH 7.41, the other contained 60 pl of a ~-[U-~~C]fructosyl-aminocaproate solution in the same buffer. The amount of nonspecific binding was evaluated by performing the assay in the presence of an excess of unlabeled fructosyl-e-aminocaproate (200 mM). After incubation at 4 "C for 20-22 h in a rotary shaker, two aliquots of 20 pl were withdrawn from each well, and the radioactivity was measured. For the inhibition studies, the '"C-labeled Amadori product was used a t 6 PM constant concentration with and without a 10-fold molar concentration of the compound to be tested. The ability ofABP to bind different compounds was expressed as percent of the binding of e-[U-"Clfructosyl-aminocaproate. For the saturation studied, the "Clabeled Amadori product was used at concentration ranging from 0.5 to 18 p~. A plot of bound Amadori versus concentration was obtained, and the binding parameters were calculated according to the Michaelis-Menten equation using nonlinear fitting analysis.
Enzyme-linked Immunoassay (ELISA) for Binding to Glycated Proteins-For investigation of the ability of ABP to bind to glycated proteins, an ELISA sandwich assay was performed. Briefly, glycated proteins (BSA and RNase) were immobilized on microtiter plates, and after blocking with unglycated BSA, ABP (10-100 ng) was added to each well. Following 16 h of incubation at 4 "C, bound ABP was detected by ELISA using a monoclonal antibody to ABP (see below) and goat antimouse IgG conjugated with alkaline phosphatase.
Production of a Monoclonal Antibody to ABP-To investigate the ABP binding to glycated proteins and for detection in other organisms, a monoclonal antibody was produced using the methods of Schulman et al. (21) and modified by Kaetzel et al. (22).
Ascites or culture supernatant was pretreated with Seroclear Reagent (Calbiochem) for delipidation, and IgG rich ascites were purified by FPLC on a protein G Superose column (Pharmacia Biotech Inc.). The specific IgG was eluted with 0.1 M HEPES-HCI, (pH 2.7). The antibody was typed by immunodiffusion in agarose gels (ICN Biochemicals) and found to belong to the class IgG,.
An ELISA consisting of 100 nglwell ABP as coating agent was developed as described previously (38). Following blocking of the 96-well plate with 200 p1 of 1 g/lOO ml ovalbumin, the plate was washed 3 times with phosphate-buffered saline-Tween 20 (0.05 g/lOO ml) and used for detection of purified antibodies, mouse sera, or culture supernatant using goat anti-mouse IgG conjugated with alkaline phosphatase and p-nitrophenyl phosphate as substrate.
SDS-Polyacrylamide Gel Electrophoresis SDS-PAGE was performed on a 12% polyacrylamide resolving gel and 4% stacking gel according to the method of Laemmli (23). Protein samples were solubilized both with reducing and nonreducing SDS sample buffer. The gels were stained for proteins with Coomassie Blue R-250 and silver stain.

Molecular Mass Measurement
The protein molecular mass was estimated on SDS-PAGE by extrapolation from the semilogarithmic plot of the relative mobility of standard proteins run simultaneously.
Protein Measurement Absorbance at 280 nm was used to monitor the protein elution from column chromatography. Protein concentration was measured by the method of Bradford (24) using BSA as standard.

Amino-terminal Sequence Analysis
A partial sequence of the protein was obtained according to Matsudaira (25). Briefly, the purified protein was subjected to SDS-PAGE and electroblotted on Immobilon membrane using 1 mM CAPS, 10% methanol buffer (pH 11). After staining with Coomassie Blue, the blot was air dried, and the protein band was cut out. Sequence analysis was per- formed at the Molecular Biology Core Laboratory (Department of Biochemistry, Case Western Reserve University) with an Applied Biosystems 477A gas-phase sequencer equipped with an on-line analyzer for phenylthiohydantoin-derivatized amino acids. The NH,-terminal sequence was entered in the FASTA program to search the GenBank Sequence Data base (26).

RESULTS
Strain Selection-A mucoid and mobile P. aeruginosa soil organism was found to utilize the Amadori product as sole carbon source. The organism was isolated from soil samples by direct selection and combination of growth in broth and agar. Typing was performed at the Microbiology Laboratory of the Institute of Pathology. In order to investigate whether the growth of the isolated microorganism was due to the utilization of the Amadori product itself or to spontaneous conversion of the product to free glucose and eaminocaproate, the growth rates of the soil strain with various substrates were compared with those of a control strain of P . aeruginosa (ATCC 27853).
While both strains could grow on minimal medium supplemented with glucose and eaminocaproic acid, only the soil strain grew on medium containing glycated-eaminocaproate, and none of the strains was capable of growth on glucose or aminocaproic acid only (not shown).
Isolation of the MP-To search for protein(s) involved in recognition of the Amadori product, affinity chromatography of the bacterial extract was carried out on glycated and unglycated lysine-Sepharose as described, and the eluate was analyzed by SDS-PAGE. A single protein was found in the soil strain extract upon elution from glycated lysine-Sepharose. However, no protein band was obtained from the extract passed over unmodified affinity substrate (Fig. lA, lanes 3 and 41,

Amadori Binding Protein
suggesting that the identified protein binds to the fructosyllysine adduct of the affinity support rather than nonspecifically to the affinity support. The protein baptized ABP was not found in the extracts from Pseudomonas (Fig. L4, lane 6) and E. coli control strains (not shown), both unable to utilize the Amadori product. Moreover, no protein was eluted from the glycated lysine-Sepharose column when the same affinity experiment was performed with cell extract from soil strain cultivated in minimal medium supplemented with glucose and 6-aminocaproic acid or in nutrient broth (data not shown), suggesting that the synthesis of ABP is related to the presence of the Amadori product in culture medium. To confirm this hypothesis, an induction experiment was performed in which the soil strain-washed cells grown in nutrient broth culture were incubated in minimal medium containing efructosyl-aminocaproate. As shown in Binding Studies-To evaluate the relationship between the extent of glycation of lysine-Sepharose and the binding of the purified protein, the binding was studied by varying the extent of glycation of the affinity substrate. A strong correlation was found between the amount of protein bound and the extent of glycation of lysine-Sepharose (not shown).
The binding specificity of ABP for the Amadori product was investigated by affinity chromatography on various substrates. The protein was not retained by unmodified lysine-Sepharose and NaBH,-reduced glycated lysine-Sepharose (Fig. 2, lanes 3  and 4 ) . Moreover, the binding to the glycated substrate was not affected by carrying out the glycation reaction with EDTA (Fig.  2, lane 51, excluding, thereby, the binding to carboxymethyllysine, a degradation product of the Amadori compound that forms in presence of oxygen and metal ions (27).
Binding specificity was also studied by measuring the ability of unlabeled compounds to compete with the binding of [U-'4C]fructosyl-~-aminocaproate in the equilibrium dialysis radioligand binding assay (Table I). Of the compounds tested, only the Amadori products of glucose were strong competitive inhibitors but not the one of ribose (6-ribosyl-lysine). Both Eand a-fructosyllysine were inhibitors; however, the affinity of ABP for the a-amine derivative was half of that for efructosyllysine. Complete inhibition of the ABP binding to [U-14Clfructosyl-eaminocaproate was observed also with proteinasedigested glycated proteins. Free amines and sugars, as well as carboxymethyllysine showed no or very low inhibitory effect. Surprisingly, when the binding assay was performed with NaBH,-reduced fructosyl-E-aminocaproate, 75% inhibition was observed. This is in contrast with the data from binding to borohydride-reduced glycated lysine-Sepharose and with the absence of inhibition with methylglucamine, a model compound of NaBH,-reduced Amadori product.
When the binding of glycated-eaminocaproate to ABP was investigated by increasing concentration of the ligand, saturation was observed (Fig. 3)

TABLE I Inhibition of [U-'4C]fructosyl-e-aminocaproate binding to ABP by various compounds
Results are expressed as percentage of the binding of I4C-labeled fructosyl-e-aminocaproate. Glycated and unglycated BSA and RNase tion of enzymatic digestion was confirmed by SDS-PAGE. Proteinase k were digested with proteinase k (enzyme to protein ratio, 1:lO). Complewas removed by ultrafiltration (Centricon 10, Amicon Inc.), and the digested glycated proteins were diluted to a final furosine concentration of 60 J~M . Digested control proteins were diluted to the same protein concentration of the glycated one (about 3 mg/ml). All other inhibitors were tested a t 60 J~M concentrate. The data represent means of two or more experiments. A monoclonal antibody to ABP was produced for determination of ABP binding to glycated proteins and for evaluation of its presence in other microorganisms, including yeast and rat liver homogenate. The antibody was able to bind ABP as demonstrated by ELISA (Fig. 4). Western blotting revealed that ABP was present in the Pseudomonas soil organism grown in Amadori product, but absent in control strain as well as in several Gram-positive and -negative bacteria, yeast, and rat liver extract.
The ability ofABP to bind to glycated proteins immobilized in microtiter plates was investigated using an ELISA sandwich assay with the monoclonal anti-ABP antibody. Proteins (BSA, RNase) were reacted with D-glucose up to 30 days, and a time course experiment was performed. However, no ABP binding was detected as a function of glycation (not shown). DISCUSSION A novel protein that specifically binds Amadori products originating from glucose and low molecular weight amines was isolated from a soil strain of Pseudomonas sp., which was selected on the basis of its ability to utilize a synthetic Amadori product as the only carbon source to grow. ABP was isolated by a one-step chromatography on glycated lysine-Sepharose and was found to be highly specific in its substrate requirements, to the extent that only Amadori products originating from glucose could compete with glycate eaminocaproate for its binding to ABP. This finding suggests that the ABP-ligand interaction is predominantly mediated by the sugar-derived portion of the Amadori product. The fact that the protein was not retained by glycated affinity substrate after NaBH, reduction further suggested that the Amadori product should be in the hemiketal configuration in order to be recognized by the binding protein.
In contrast to the findings above, investigation of ABP binding properties by equilibrium dialysis revealed that the NaBH,reduced fructosyl-e-aminocaproate was able to compete with the binding of the unreduced Amadori product to ABP by 75% (Table I). However, methylglucamine, a compound identical with borohydride-reduced glycated methylamine did not inhibit the binding of fructosyl-eaminocaproate to ABP, thereby raising the question of how to reconcile these observations. Carboxymethyl-lysine did not appear to be responsible for ABP binding specificity since it was not retained by the affinity substrate when glycation of lysine-Sepharose was carried out under deaerated conditions in presence of EDTA, thereby excluding the possibility that carboxymethyllysine instead of the Amadori compound was the actual ligand ( binding of fructosyl-E-aminocaproate by only 30%. These data, suggest that ABP recognizes both the carbohydrate and the alkyl side chain of the Amadori product in the region around the ketoamine bond, with the sugar portion, however, being predominant. The ability of ABP to bind to glycated lysine-Sepharose and enzymatically digested glycated proteins but not to intact glycated proteins suggests presence of steric inhibition. The latter was apparently absent in lysine-Sepharose, possibly due to the spacer arm linking the a-amino group of L-lysine to the beads. The high affinity of AF3P for free ligands and the decreased affinity for protein-bound Amadori product is reminiscent of similar behavior of antihapten antibodies (38,39).
While this study was in progress, we became aware of the work of Horiuchi (311, which described the purification and properties of a fmctosyl-amino acid oxidase from a soil strain of Corynebacterium sp. This enzyme, of molecular mass 44 kDa,

Amadori Binding Protein
decomposes Amadori products of a-L-amino acid to a-ketoaldehydes (glucosone) and a-L-amino acids under oxidative condi-  . (33), although the latter enzyme was not isolated. In both cases, the selection of the microorganism from the soil samples was performed using an Amadori product as sole carbon source in the culture medium. The similarity of the molecular mass of a-amino acid oxidase and ABP, and the fact that Pseudomonas sp. utilize glucose oxidatively suggested that both proteins could have been identical. This possibility was, however, excluded by the negative results obtained from experiments to investigate the potential enzymatic activity of ABP. Moreover, immunoblotting experiments performed with fmctosyl-amino acid oxidase generously donated by Dr. Horiuchi showed absence of cross-reactivity between ABP and the enzyme. The discovery of an Amadori binding protein in the soluble extract of a Pseudomonas soil strain grown on Amadori product as sole carbon source, raises the question of what physiological function it could have. Since we found membrane-bound enzyme activity toward the same ligand (40), it could be that ABP is a soluble periplasmic permease protein with transport function. Although yet to be proven, this hypothesis is supported by the binding data for low molecular weight compounds and the substrate-dependent regulation of protein synthesis. A variety of permeases from Pseudomonas and other bacteria are reported to be inducible proteins. The regulation of binding protein synthesis by substrate induction is well known, especially for the permease involved in the uptake of sugar (34). Moreover, the affinity of ABP for its specific substrate is in the same molar range as reported in the literature for other bacterial binding proteins (28, 29) and for glucose uptake by intact cells (30). However, further studies will be necessary to confirm the permease activity of ABP and to investigate the relationship between Amadori product uptake and utilization. Such information will have direct relevance for cellular processing of glycated amines and proteins.
Recently, two studies have uncovered the presence of cellular receptors for glycated proteins. Krantz et al. (35) described the saturable binding of glycated proteins to macrophages that was inhibitable by coincubation with glycated €-amino lysine. Similar findings have been made by Ku and Cohen (36) in aortic endothelial cells. Thus, mammalian cells have the ability to bind and process glycated proteins. It is currently unknown whether mammalian cells secrete proteins that can bind glycated low molecular weight amines. However, studies by Erbersdobler et al. (37) have shown that free glycated lysine is absorbed, taken up by the liver, as well as excreted, suggesting that some proteins may be involved in transporting glycated amino acids across membranes.