Renal uptake of myoglobin is mediated by the endocytic receptors ...

Am J Physiol Renal Physiol 285: F451–F458, 2003. First published April 29, 2003; 10.1152/ajprenal.00062.2003.

Renal uptake of myoglobin is mediated by the endocytic receptors megalin and cubilin Jakub Gburek,1,2 Henrik Birn,1 Pierre J. Verroust,3 Bogusława Goj,2 Christian Jacobsen,4 Søren K. Moestrup,4 Thomas E. Willnow,5 and Erik I. Christensen1 Departments of 1Cell Biology and 4Medical Biochemistry, University of Aarhus, DK-8000 Aarhus C, Denmark; 2 Department of Biochemistry, Faculty of Pharmacy, Wrocław Medical University, 50139 Wrocław, Poland; 3 Institut Nationale de la Sante´ et de la Recherche Medicale´ Unite´ 538, Centre Hospitalier Universitaire, St. Antoine, 75012 Paris, France; and 5Max Delbrueck Center for Molecular Medicine, D-13125 Berlin, Germany Submitted 14 February 2003; accepted in final form 16 April 2003

Gburek, Jakub, Henrik Birn, Pierre J. Verroust, Bogusława Goj, Christian Jacobsen, Søren K. Moestrup, Thomas E. Willnow, and Erik I. Christensen. Renal uptake of myoglobin is mediated by the endocytic receptors megalin and cubilin. Am J Physiol Renal Physiol 285: F451–F458, 2003. First published April 29, 2003; 10.1152/ ajprenal.00062.2003.—Nephrotoxicity of myoglobin is well recognized as playing a part in the development of acute renal failure in settings of myoglobinuria. However, the molecular mechanism of myoglobin uptake in renal proximal tubules has not been clarified. Here, we report that the endocytic receptors megalin and cubilin are involved in renal reabsorption of myoglobin. Both receptors were captured from solubilized renal brush-border membranes by affinity chromatography using myoglobin-Sepharose. Myoglobin bound to purified megalin and cubilin with Kd values of 2.0 and 3 ␮M, respectively, as evaluated by surface plasmon resonance analysis. Apomyoglobin bound to megalin with the same affinity, and the affinity of apomyoglobin to cubilin was reduced (Kd ⫽ 5 ␮M). Radioiodinated myoglobin could be displaced by apomyoglobin in inhibition studies using isolated renal brush-border membranes (Ki ⬃ 2 ␮M). Receptorassociated protein as well as antibodies directed against megalin and cubilin markedly inhibited the uptake of fluorescent-labeled myoglobin by cultured yolk sac BN-16 cells. The significance of megalin- and cubilin-mediated endocytosis for myoglobin uptake in vivo was demonstrated by use of kidney-specific megalin knockout mice. Injected myoglobin was extensively reabsorbed by megalin-expressing proximal tubular cells, whereas there was very little uptake in the megalin-deficient cells. In conclusion, this study establishes the molecular mechanism of myoglobin uptake in the renal proximal tubule involving the endocytic receptors megalin and cubilin. Identification of the receptors for tubular uptake of myoglobin may be essential for development of new therapeutic strategies for myoglobinuric acute renal failure.

frequently develops in clinical settings associated with increased urinary excretion of myoglobin. Myoglobinuria is a consequence of severe muscle injury, termed rhabdomyolysis, which is usually associated with trauma but may occur in variety of

other clinical conditions such as hyperthermia, muscle ischemia, or exposure to certain toxins or drugs (10, 24). Dependent on etiology myoglobinuria may be accompanied by additional complications but seemingly deposition of myoglobin in the kidney is a principal factor contributing to the decline of renal function. At the kidney level, the main pathological events in the genesis of myoglobinuric ARF have been identified as intense renal vasoconstriction, tubular obstruction by intraluminal pigment casts, and necrosis of proximal tubular epithelium due to direct intra- and/or extracellular toxicity of myoglobin. Because there is much evidence for the occurrence of oxidative damage to the kidney in rhabdomyolysis, the toxicity of myoglobin has been attributed to catalysis of free radical reactions. The precise mechanism, whether due to free iron, heme, or heme protein, and which free radical mechanism is involved, as well as which cellular organelles are affected, are still debated (11, 26). It has been convincingly shown that myoglobin undergoes reabsorption from the glomerular filtrate and is catabolized within proximal tubule cells (2, 14, 21). Thus it is conceivable that particular components of myoglobin can act on different intracellular levels with the involvement of diverse radical species during reabsorption of holoprotein and its subsequent decomposition. Consequently, it seems to be of great interest to elucidate the molecular mechanism of myoglobin uptake and degradation in the kidney. Several glomerular filtrate proteins are reabsorbed as a complex with megalin and/or cubilin, the multiligand endocytic receptors residing in the membrane of proximal tubular epithelial cells. Megalin is a 600-kDa type I transmembrane glycoprotein belonging to the LDL receptor family, whereas cubilin is a 460-kDa glycoprotein that lacks a classic transmembrane domain and has no homology to any known endocytic receptors. As the vast extracellular domains of these receptors can accommodate a variety of ligands, and the receptors can act both independently or following

Address for reprint requests and other correspondence: J. Gburek, Dept. of Cell Biology, Institute of Anatomy, Wilhelm Meyers Alle, Bldg. 234, Aarhus University, DK-8000 Aarhus C, Denmark (E-mail: [email protected]).

The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked ‘‘advertisement’’ in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

endocytosis; acute renal failure

ACUTE RENAL FAILURE (ARF)

http://www.ajprenal.org

0363-6127/03 $5.00 Copyright © 2003 the American Physiological Society

F451

F452

CLEARANCE RECEPTORS FOR MYOGLOBIN IN THE KIDNEY

association as a dual-receptor complex, it seems plausible that they facilitate uptake of most proteins from the primary filtrate (7). This is supported by the finding of low-molecular-weight proteinuria in megalin knockout mice and in dogs that bear an inherited disorder of intracellular cubilin processing (16, 25). Furthermore, low-molecular-weight proteinuria also develops in patients who suffer from Imerslund-Gra¨ sback syndrome, a rare autosomal disorder caused by mutations in the cubilin gene (1). We have recently shown that megalin and cubilin are responsible for renal reabsorption of hemoglobin, a heme protein structurally related to myoglobin (13). Thus here we aim at resolving the role of those receptors in myoglobin reabsorption. MATERIALS AND METHODS

Proteins. Horse myoglobin was obtained from Serva Electrophoresis (Heidelberg, Germany), and this preparation was used throughout the study. BSA and ovoalbumin were purchased from Sigma (St. Louis, MO). Recombinant receptorassociated protein (RAP) was prepared as described elsewhere (20). Apomyoglobin was prepared from myoglobin using the acid acetone procedure (22). Fluorescence-labeled myoglobin was synthetized by coupling 5(6)-carboxyfluorescein-N-hydroxysuccinimide ester (FLUOS; Roche Diagnostics, Mannheim, Germany) to myoglobin amino groups using a fluorescein-labeling kit according to the manufacturer’s instructions. Briefly, 5 mg of myoglobin dissolved in PBS, pH 7.4, were incubated with 1.2 mg FLUOS for 2 h at room temperature with gentle mixing. Unbound FLUOS was removed by Sephadex G-25 gel filtration. The approximate fluorescein-to-protein ratio of the preparation was 10. Aliquots of FLUOS-myoglobin (1 mg/ml) were stored at ⫺20°C until used. 125 I-myoglobin was prepared by use of Iodo-gen according to Salacinski et al. (23). Specific activities of the tracer preparations were in the range of 1.0–1.5 ␮Ci/␮g protein. Protein concentrations were determined using a protein assay reagent (Pierce, Rockford, IL). Antibodies. Affinity-isolated IgG to horse myoglobin was purchased from Bethyl Laboratories (Montgomery, TX). Sheep anti-rat megalin and rabbit anti-rat cubilin antisera were obtained as described previously (15, 19). Sera IgG fractions were prepared by protein A-agarose affinity chromatography according to the manufacturer’s instructions (Pierce). A nonimmune sheep serum IgG fraction was obtained from Sigma. Alexa-conjugated secondary antibodies were purchased from Molecular Probes (Eugene, OR). All other antibodies used in this study were purchased from Dako (Glostrup, Denmark). Purification of myoglobin receptors by affinity chromatography. Rat renal brush-border membranes were prepared and solubilized using Triton X-100 as previously described (12). The membrane protein extract was recirculated at 0.2 ml/min flow through a 1.5-ml rat myoglobin-Sepharose column equilibrated with PBS, pH 7.4, 0.6 mM CaCl2, and 0.5% 3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonate (CHAPS). The column was washed with 30 ml PBS, pH 7.4, 0.6 mM CaCl2, 0.5% CHAPS, and 30 ml of the same buffer containing 0.5 M NaCl, and again with 10 ml of the first buffer. Bound proteins were eluted with PBS, pH 5.0, 10 mM EDTA, and 0.5% CHAPS. Collected 1-ml fractions were concentrated 10 times by ultrafiltration using Centricon YM 10 (Millipore, Bedford, MA) and analyzed under nonreducing AJP-Renal Physiol • VOL

conditions by 4–16% SDS-PAGE. Protein bands were visualized by Gelcode blue stain reagent (Pierce). For immunochemical analysis, proteins were blotted onto Hybond ECL nitrocellulose membranes (Amersham Pharmacia Biotech, Little Chalfont, Buckingamshire, UK). Nitrocellulose membranes were blocked by 5% skim milk in 80 mM Na2HPO4, 20 mM NaH2PO4, and 0.1% Tween 20, pH 7.5 (PBS-T) for 1 h and incubated with primary antibody in PBS-T overnight at 4°C. After being washed with PBS-T, the blots were incubated with horseradish peroxidase-conjugated secondary antibody diluted 1:3,000 in PBS-T. ECL-PLUS reagent (Amersham Pharmacia Biotech) and the Fluor-s imaging system (Bio-Rad Laboratories, Hercules, CA) were used for chemiluminescent visualization. Kinetics of myoglobin and apomyoglobin binding to cubilin and megalin. The binding of myoglobin and apomyoglobin to megalin and cubilin was studied by surface plasmon resonance analysis on a BiaCore 2000 instrument (BiaCore, Uppsala, Sweden). The procedure was essentially as described previously (5). Briefly, BiaCore type CM5 sensor chips were activated with a 1:1 mixture of 0.2 M N-ethyl-N⬘-(3-dimethylaminopropyl)carbodiimide and 0.05 M N-hydroxysuccimide in water according to the manufacturer’s recommendations. Megalin and cubilin were purified by RAP or IF-B12 affinity chromatography, respectively. The preparations produced single bands in SDS-PAGE followed by Coomassie brilliant blue staining. No cross-contamination of the two proteins could be detected. The proteins were immobilized at concentrations up to 50 ␮g/ml in 10 mM sodium acetate, pH 4.5, and the remaining binding sites were blocked with 1 M ethanolamine, pH 8.5. The resulting receptor densities were in the range of 23–40 fmol receptor/mm2. A control flow cell was made by performing the activation and blocking procedures only. Immobilized receptor proteins were reduced by injection of 0.5% dithiothreitol in 6 M guanidine hydrochloride, 5 mM EDTA, and 50 mM Tris, pH 8.0, into the flow cell. Samples were dissolved in 10 mM HEPES, 150 mM NaCl, 2 mM CaCl2, and 0.005% Tween 20, pH 7.4, or 10 mM HEPES, 150 mM NaCl, 20 mM EGTA, and 0.005% Tween 20, pH 7.4. Sample and running buffers were identical. The regeneration of sensor chips after each analysis cycle was performed with 1.6 M glycine-HCl buffer, pH 3.0. The BiaCore response is expressed in relative response units, i.e., the difference in response between proteins and the control flow channel. Kinetic parameters were determined by using BIAevaluation 3.1 software. Competition between myoglobin and apomyoglobin for binding to receptor sites was studied by the rapid filtration technique using isolated rat brush-border membranes as previously described (12). In brief, membranes (100 ␮g protein) were incubated in binding buffer (Tris-buffered saline) with tracer amounts of 125I-myoglobin (⬃104 counts/min) and increasing concentrations of apomyoglobin in the range of 0.3–5 ␮M in a final volume of 0.2 ml for 60 min. After incubation, 0.15-ml samples were applied onto GVWP 0.22-␮m membrane filters (Millipore) and washed with 5 ml of binding buffer to remove unbound ligands. The radioactivity of the filters corresponding to the amount of bound 125Imyoglobin was measured in a gamma counter (Polon). Ki was evaluated by computerized nonlinear regression analysis using Prizm software (GrapPad Software). Immunohistochemistry. Immunohistochemical studies were performed in kidneys excised from control mice (C57BL) or kidney-specific megalin knockout mice (15) injected with myoglobin (35 mg/kg body wt in PBS, pH 7.4) into the femoral vein at 15 and 30 min after injection, respectively.

285 • SEPTEMBER 2003 •

www.ajprenal.org


Cortical tissue specimens were prepared from kidneys after fixation with 2% paraformaldehyde in 0.1 M cacodylate buffer, pH 7.4, by retrograde perfusion through the abdominal aorta. Blocks of tissue were further fixed by immersion in the same solution for 1 h and transferred to 2.3 M sucrose in PBS, pH 7.4, for 0.5 h before being frozen in liquid nitrogen. Semithin cryosections (0.8 ␮m) were cut using a Reichert Ultracut S microtome (Richert-Jung, Vienna, Austria) and placed onto glass slides. Endogenous peroxidase activity was quenched with PBS, pH 7.4, 10% methanol, and 3% H2O2, and nonspecific binding was blocked with PBS, pH 7.4, 1% BSA, and 0.05 M glycine. For the immunoperoxidase reaction, the sections were subsequently incubated with goat anti-myoglobin affinitypurified IgGs diluted 1:400–1:800 and peroxidase-conjugated rabbit IgG anti-goat IgGs diluted 1:300. The reaction was visualized with diaminobenzidine. The sections were counterstained with Meyer’s hematoxilin. In immunofluorescence studies, the following antibodies were used: goat anti-horse myoglobin affinity-purified IgGs diluted 1:400–1:800; sheep anti-rat megalin antiserum diluted 1:25,000–1:50,000; rabbit anti-rat cubilin antiserum diluted 1:2,000–1:4,000; Alexa 568-labeled donkey IgG antigoat IgGs diluted 1:300; Alexa 488-labeled donkey IgG antisheep IgGs diluted 1:300; and FITC-conjugated swine IgG anti-rabbit IgGs diluted 1:40. The sections were analyzed using Leica SP2 confocal microscope. Uptake studies in cell culture. Rat yolk sac carcinoma BN-16 cells (18) were routinely grown in 25-cm2 plastic culture flasks (Corning Costar, Badhoevedrop, Holland) in Eagle’s MEM (Bio-Whittaker, Walkersville, MD) supplemented with 10% fetal calf serum (Biological Industries, Fredensborg, Denmark), 2 mM L-glutamine, 50 U/ml penicillin, and 50 ␮g/ml streptomycin (Bio-Whittaker) in a humidified atmosphere containing 5% CO2 at 37°C. Cells were passaged every fourth day at a split ratio of 1:5 by trypsinization with 500 mg/l trypsin and 200 mg/l EDTA (Bio-Whittaker). For uptake experiments, cells were cultured in eightchamber glass slides (Nalge Nunc International, Naperville, IL). One day before the cells reached confluence, the medium was replaced with medium supplemented with 0.5% ovoalbumin instead of 10% serum (serum-free medium). Cell monolayers were incubated with 5 ␮g/ml FLUOS-myoglobin in serum-free medium for 10 min and fixed with 1% paraformaldehyde in PBS, pH 7.4, for 3 min. For inhibition studies,

F453

the following proteins were added to the incubation mixture: 1 ␮M recombinant receptor-associated protein, a 200-mg/l sheep anti-rat megalin IgG serum fraction and a sheep nonimmune IgG serum fraction, or 400-mg/l rabbit anti-rat cubilin IgG serum fraction and rabbit nonimmune IgG serum fraction. The slides were mounted with 50% glycerol, 2% N-propyl-gallat, and 2.4% Tris and examined using a fluorescence microscope (Leica DMR) equipped with a color video camera (Sony 3CCD). RESULTS

Myoglobin affinity chromatography. To identify potential endocytic receptors for myoglobin in the kidney proximal tubule, we performed affinity chromatography on a myoglobin-Sepharose column. Solubilized rat renal brush-border membranes were passed over the affinity matrix, and, after an extensive washing, the retained proteins were eluted at pH 5.0. Analysis of final acid eluate in SDS-PAGE revealed the presence of two proteins, which comigrated with cubilin and megalin, seen as the two predominant bands at the top area of the brush-border membrane fraction lane (Fig. 1). We further confirmed the identity of the proteins by Western blot analysis of the peak fraction using specific antibodies to megalin and cubilin (Fig. 1). Surface plasmon resonance analysis. Kinetic parameters of myoglobin binding to megalin or cubilin were examined by means of surface plasmon resonance analysis. Kd values for the interactions of myoglobin with megalin or cubilin were 2.0 and 3.0 ␮M, respectively (Fig. 2, A and B). Apomyoglobin bound to megalin with the same affinity as myoglobin (Kd ⫽ 2 ␮M), whereas the affinity of apomyoglobin for cubilin was lower (Kd ⫽ 5 ␮M). The binding of both myoglobin and apomyoglobin was completely abolished in the presence of EDTA (Fig. 2, C and D). Competition of both myoglobin forms for binding sites on the receptors was tested in an inhibition study using isolated brushborder membranes. Radioiodinated myoglobin could be displaced by apomyoglobin with a Ki of 2 ␮M (Fig. 3). Uptake of myoglobin in vitro. The role of megalin and cubilin in the endocytic uptake of myoglobin was in-

Fig. 1. Analysis of affinity-purified myoglobin-binding proteins by SDS-PAGE and Western blotting. Rat renal brush-border membranes solubilized in 1% Triton X-100 were passed over myoglobin-Sepharose. After being washed, the bound fraction was eluted with PBS, pH 5.0, 10 mM EDTA, and 0.5% 3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonate. Left: SDSPAGE of consecutive eluate fractions (lanes 1–7) with Coomassie brilliant blue (CBB) staining. The myoglobin-binding fraction contains similar amounts of cubilin and megalin. Mb, 1 ␮g myoglobin; BB, brush-border membranes (10 ␮g protein); M, molecular mass standards. Right: Western blotting of peak fraction (lane 4) from myoglobin-Sepharose affinity chromatography using anti-cubilin or anti-megalin antibodies for immunoperoxidase chemiluminescent visualization.

AJP-Renal Physiol • VOL


www.ajprenal.org

F454


Fig. 2. Surface plasmon resonance analysis of myoglobin binding to cubilin and megalin. Megalin and cubilin were immobilized onto sensor chips, and myoglobin samples (40 ␮l) were passed over the flow cells at 5 ␮l/min. A and B: sensorgrams with myoglobin concentrations in the range of 0.2–5 ␮M were recorded for evaluation of kinetic parameters. The affinity of myoglobin for megalin (A) was determined as Kd ⫽ 2.0 ␮M and the affinity of myoglobin for cubilin as Kd ⫽ 3.0 ␮M, assuming 1 class of binding sites for each receptor. C and D: sensorgrams with myoglobin (solid lines) and apomyoglobin (dotted lines) in a concentration of 5 ␮M. The binding of myoglobin as well as apomyoglobin was almost abolished (below 5 response units) when 20 mM EDTA was included in sample and running buffer, showing calcium dependency of the interactions.

vestigated by an inhibition study in cultured BN-16 cells, a cell line derived from the rat yolk sac. The cells could intensively internalize FLUOS-myoglobin from the incubation medium, which appeared as green fluorescence accumulating within the endosomal/lysosomal compartment of the cell. The uptake was site limited because an excess of unlabeled myoglobin (20 ␮M) virtually prevented fluorescent labeling (not shown). We observed a marked inhibition of FLUOSmyoglobin uptake with anti-megalin or anti-cubilin antibodies at a concentration of 200 and 400 ␮g/ml, respectively, or with 1 ␮M RAP (Fig. 4).

Fig. 3. Competition of myoglobin and apomyoglobin for receptorbinding sites. Isolated brush-border membranes were incubated with tracer amount of radioiodinated myoglobin and increasing concentrations of apomyoglobin in the range of 0.3–5.0 ␮M for 1 h. The total amount of bound (B) radioligand was determined by the rapid filtration technique. Apomyoglobin inhibited myoglobin binding to receptor sites with a Ki of ⬃2 ␮M. AJP-Renal Physiol • VOL

Uptake of myoglobin in vivo. The significance of megalin- and cubilin-mediated endocytosis for myoglobin reabsorption in vivo was investigated by immunocytochemistry in kidney sections from normal and kidney-specific megalin knockout mice (15). Because endogenous myoglobin could be detected in neither normal mouse nor knockout mouse proximal tubule (data not shown), we performed our experiments using mice injected with exogenous myoglobin at a dose of 35 mg/kg body wt. Thirty minutes after injection, a strong immunoperoxidase reaction with anti-myoglobin antibodies was identified mainly along the brush border and within the apical endosomal/lysosomal compartment in virtually all cells of the proximal tubule from normal mice. There was also additional immunoreactivity in the extratubular space (Fig. 5A). In the megalin knockout mouse, some proximal tubular cells were devoid of immunoreactivity, whereas in others the distribution of the immunoreaction was analogous to that in the normal mouse (Fig. 5B). To demonstrate a relationship between the expression of megalin or cubilin and reabsorption of myoglobin in those two subsets of cells, we employed the double-immunofluorescence technique. Normal and deficient cells could be easily distinguished by anti-megalin antibodies. There was a pronounced reduction of cubilin expression in the cells lacking megalin. Both receptors colocalized mainly in the apical membrane area (Fig. 6). Megalin-immunoreactive cells exhibited substantial deposition of myoglobin at the brush border and in the apical endosomal/ lysosomal apparatus as well as in vesicular structures located deeper in the cell. Myoglobin and megalin or cubilin, respectively, colocalized at the brush border and in vesicles adjacent to the apical plasma mem-


www.ajprenal.org


F455

Fig. 4. Characterization of myoglobin uptake by BN-16 cells using fluorescence microscopy. Cells were grown to confluence on glass slides and incubated with 5 ␮g/ml 5(6)-carboxyfluorescein-N-hydroxysuccinimide ester myoglobin in Eagle’s MEM medium containing 0.5% ovoalbumin for 10 min alone (A) or in the presence of 1 ␮M receptor-associated protein (RAP; B), 200 mg/l nonimmune sheep IgG (C), 200 mg/l sheep anti-megalin IgG (D), 400 mg/l nonimmune rabbit IgG (E), or 400 mg/l rabbit anti-cubilin IgG (F). The uptake of myoglobin is markedly inhibited by RAP and anti-megalin or cubilin antibodies (compare A and B, C and D, E and F). Magnification: ⫻700.

Fig. 5. Immunoperoxidase demonstration of myoglobin uptake in normal and kidney-specific megalin knockout mouse. The animals were injected with myoglobin (35 mg/kg body wt), perfusion fixed after 30 min, and the kidneys were processed for cryoimmunocytochemistry. A: cortex of normal mouse. B: cortex of kidney-specific megalin knockout mouse. Note that all proximal tubular cells are labeled in the normal mouse, whereas in megalin knockout mouse a number of cells remined unlabeled (*). Arrows, brush-border staining; arrowheads, vacuolar staining. Magnification: ⫻1,200. AJP-Renal Physiol • VOL


www.ajprenal.org

F456


Fig. 6. Identification of myoglobin uptake in kidney-specific megalin knockout mice by double immunofluorescence. The animals were injected with myoglobin (35 mg/kg body wt), perfusion fixed after 30 min, and the kidneys were processed for cryoimmunocytochemistry. A–C: expression of megalin and cubilin in the proximal tubule. D–F: colocalization of megalin and injected myoglobin in the proximal tubule. G–I: colocalization of cubilin and injected myoglobin in the proximal tubule. Note that in megalin-deficient cells, expression of cubilin is markedly reduced and the reabsorption of myoglobin is negligible compared with megalin-expressing cells. Endocytosis of myoglobin by megalin-expressing cells is reflected by its colocalization with megalin and cubilin at the brush border and in the apical endosomal compartment. Magnification: ⫻1,200.

brane. Uptake of myoglobin by the cells devoid of megalin was either very low or not seen at all. DISCUSSION

Megalin and cubilin are the major endocytic receptors responsible for reabsorption of glomerular ultrafiltrate proteins in the proximal tubule. Here, we report that the receptors also facilitate epithelial uptake of myoglobin, a highly nephrotoxic protein released into the circulation during rhabdomyolysis. This is supported by several lines of evidence. Affinity chromatography of solubilized renal brush-border membranes on a myoglobin matrix yielded principally megalin and cubilin in approximately equal amounts. AJP-Renal Physiol • VOL

Lack of other proteins in the affinity preparation suggested that the receptors are solely responsible for myoglobin reabsorption in the proximal tubule. Because purified megalin and cubilin retained myoglobinbinding activity, we could characterize the kinetics of the binding (interactions) using surface plasmon resonance analysis. Based on Kd values, 2 ␮M for megalin and 3 ␮M for cubilin, the interactions can be classified as “low-affinity binding.” Low affinity might raise the question of the efficiency of scavenging and thereby the pathophysiological significance of such interactions. However, in the case of the kidney proximal tubule, the low affinity is likely to be compensated for by copious amounts of the receptors at the brush border and


www.ajprenal.org


increased glomerular filtrate mass convection facilitated by the microvillar apparatus. Indeed, megalinand cubilin-ligand interactions of the above strengths turned out to be essential for the gross uptake of other ligands such as albumin or hemoglobin (4, 13). Furthermore, the plasma concentration of myoglobin in most common rhabdomyolytic states is rather high, exceeding the estimated Kd values severalfold (3). To clarify whether these relatively weak interactions can genuinely promote the uptake of myoglobin, we employed a cell culture model in our studies. Because in cultured cells of proximal tubule origin like LLCPK1 or opossum kidney cells the expression of megalin and cubilin is much lower compared with those in vivo, we chose BN-16 cells for these experiments. BN-16 cells, originating from yolk sac epithelium, are structurally and functionally similar to renal proximal tubule cells but exhibit high rates of cubilin- and megalin-mediated endocytosis. These cells have previously been used as an in vitro model of proximal tubular endocytosis (17, 18). Significantly, accumulation of fluorescence-labeled myoglobin in the cells was inhibited by antibodies raised against purified receptors and by RAP, a chaperone that affects binding of most megalin and cubilin ligands (6). To assess the role of the receptors in the clearance of myoglobin in vivo, we performed experiments using kidney-specific megalin knockout mice. The mice were designed specifically to conquer the problems of high perinatal lethality and severe complex phenotype associated with full megalin gene knockout, so far limiting the usefulness of the model. Renal specific knockout mice present a severe tubular reabsorption deficiency associated with up to a 90% reduction in the number of proximal tubular cells expressing megalin but exhibit normal development and viability (15). Mice were injected intravenously with myoglobin at a dose of 35 mg/kg, which is compatible with the concentration range observed in rhabdomyolysis (3). A comparison of myoglobin deposition in megalin-deficient and normal cells after intravenous administration of the protein clearly implied that megalin-mediated endocytosis is a predominant route of myoglobin entry into proximal tubule cells. In normal cells, myoglobin colocalized with the receptors at the brush border and within the apical vacuolar apparatus, which reflected its binding to the receptors and subsequent endocytosis. Part of the endocytosed myoglobin could be detected in vacuolar structures devoid of the receptors, which is consistent with early sorting of the receptors into the recycling compartment (8). Deposition of injected myoglobin in megalin-deficient cells was strikingly reduced and limited to relatively small vacuolar structures, which resembled uptake by fluid-phase endocytosis. This process has been previously characterized in the proximal tubule as a minor nonreceptor endocytic transport system accounting for ⬍2% of the total uptake (9). Besides the epithelium, there was also deposition of myoglobin in the interstitium, indicating a substantial leak of the protein through the capillary AJP-Renal Physiol • VOL

F457

walls. The finding possibly discloses one more potential target of myoglobin toxicity in the kidney. Necrosis of proximal tubular epithelium is a trait/ histopathological phenomenon found in the myoglobinuric kidney and thus is believed to be an important factor in the development of the disease. Substantial evidence has linked this lesion to iron-driven oxidative stress in connection with intracellular breakdown of the hemoprotein (26). In light of our findings, one could envisage that restriction of megalin-mediated endocytosis of myoglobin, especially early after rhabdomyolytic insult, could have a beneficial effect on myoglobinnuric ARF. A potential competitor of myoglobin endocytosis that emerged during this study was apomyoglobin. Lacking the neprotoxic heme, apomyoglobin can displace myoglobin from the receptor-binding sites. However, it is not known to what extent such competition can occur in vivo and whether administration of the doses required for inhibition would have its own adverse effects. In conclusion, our study establishes a molecular mechanism of myoglobin uptake in the renal proximal tubule involving the endocytic receptors megalin and cubilin. Our findings may offer a perspective for experimental therapies preventing development of myoglobinuric ARF. We suggest that organ-specific megalin knockout mice may be employed in further studies on the pathophysiology of myoglobinuric ARF. Such investigations may in particular clarify the significance of tubular myoglobin reabsorption for the development of ARF as well as its role during the maintainance and recovery phase of this disease. The technical assistance of Hanne Sidelmann, Inger Kristoffersen, and Pia Kamuk Nielsen is greatly appreciated. DISCLOSURES This work has been supported by the Danish Medical Research Council, the University of Aarhus Research Foundation, the Marie Curie Fellowship of the European Community’s Improving Human Potential Program (contract no. HPMF-CT-2001–01129), the NOVONordisk Foundation, and the Biomembrane Research Centre of Aarhus University. REFERENCES 1. Aminoff M, Carter JE, Chadwick RB, Johnson C, Gra¨ sbeck R, Abdelaal MA, Broch H, Jenner LB, Verroust PJ, Moestrup SK, de la Chapelle A CA, and Krahe R. Mutations in CUBN, encoding the intrinsic factor-vitamin B12 receptor, cubilin, cause hereditary megaloblastic anaemia 1. Nat Genet 21: 309–313, 1999. 2. Anderson WA. The use of exogenous myoglobin as an ultrastructural tracer. Reabsorption and translocation of protein by the renal tubule. J Histochem Cytochem 20: 672–684, 1972. 3. Beetham R. Biochemical investigation of suspected rhabdomyolysis. Ann Clin Biochem 37: 581–587, 2000. 4. Birn H, Fyfe JC, Jacobsen C, Mounier F, Verroust PJ, Orskov H, Willnow TE, Moestrup SK, and Christensen EI. Cubilin is an albumin binding protein important for renal tubular albumin reabsorption. J Clin Invest 105: 1353–1361, 2000. 5. Birn H, Verroust PJ, Nexø E, Hager H, Jacobsen C, Christensen EI, and Moestrup SK. Characterization of an epithelial approximately 460-kDa protein that facilitates endocytosis of intrinsic factor-vitamin B12 and binds receptor-associated protein. J Biol Chem 272: 26497–26504, 1997.


www.ajprenal.org

F458


6. Birn H, Vorum H, Verroust PJ, Moestrup SK, and Christensen EI. Receptor-associated protein is important for normal processing of megalin in kidney proximal tubules. J Am Soc Nephrol 11: 191–202, 2000. 7. Christensen EI and Birn H. Megalin and cubilin: multifunctional endocytic receptors. Nat Rev Mol Cell Biol 3: 256–266, 2002. 8. Christensen EI, Birn H, Verroust P, and Moestrup SK. Membrane receptors for endocytosis in the renal proximal tubule. Int Rev Cytol 180: 237–284, 1998. 9. Christensen EI and Maunsbach AB. Effects of dextran on lysosomal ultrastructure and protein digestion in renal proximal tubule. Kidney Int 16: 301–311, 1979. 10. David WS. Myoglobinuria. Neurol Clin 18: 215–243, 2000. 11. Don BR, Rodriguez RA, and Humphreys MH. Acute renal failure associated with pigmenturia or crystal deposits. In: Diseases of the Kidney and Urinary Tract, edited by Schrier RW. Philadelphia, PA: Lippincott Williams & Wilkins, 2001, p. 1299– 1326. 12. Gburek J and Osada J. Hemoglobin binding sites on renal brush-border membranes. Biochimie 82: 1135–1142, 2000. 13. Gburek J, Verroust PJ, Willnow TE, Fyfe JC, Nowacki W, Jacobsen C, Moestrup SK, and Christensen EI. Megalin and cubilin are endocytic receptors involved in renal clearance of hemoglobin. J Am Soc Nephrol 13: 423–430, 2002. 14. Kagen LJ. Immunofluorescent demonstration of myoglobin in the kidney. Case report and review of forty-three cases of myoglobinemia and myoglobinuria identified immunologically. Am J Med 48: 649–653, 1970. 15. Leheste JR, Melsen F, Wellner M, Jansen P, Schlichting U, Renner-Muller I, Andreassen TT, Wolf E, Bachmann S, Nykjaer A, and Willnow TE. Hypocalcemia and osteopathy in mice with kidney-specific megalin gene defect. FASEB J 17: 247–249, 2003. 16. Leheste JR, Rolinski B, Vorum H, Hilpert J, Nykjaer A, Jacobsen C, Aucouturier P, Moskaug JO, Otto A, Christensen EI, and Willnow TE. Megalin knockout mice as an animal model of low molecular weight proteinuria. Am J Pathol 155: 1361–1370, 1999. 17. Le Panse S, Ayani E, Nielsen S, Ronco P, Verroust P, and Christensen EI. Internalization and recycling of glycoprotein

AJP-Renal Physiol • VOL

18.

19.

20.

21. 22. 23.

24.

25.

26.

280 in epithelial cells of yolk sac. Eur J Cell Biol 72: 257–267, 1997. Le Panse S, Galceran M, Pontillon F, Lelongt B, van de Putte M, Ronco PM, and Verroust PJ. Immunofunctional properties of a yolk sac epithelial cell line expressing two proteins gp280 and gp330 of the intermicrovillar area of proximal tubule cells: inhibition of endocytosis by the specific antibodies. Eur J Cell Biol 67: 120–129, 1995. Moestrup SK, Nielsen S, Andreasen P, Jørgensen KE, Nykjaer A, Røigaard H, Gliemann J, and Christensen EI. Epithelial glycoprotein-330 mediates endocytosis of plasminogen activator-plasminogen activator inhibitor type-1 complexes. J Biol Chem 268: 16564–16570, 1993. Nykjaer A, Petersen CM, Møller B, Jensen PH, Moestrup SK, Holtet TL, Etzerodt M, Thøgersen HC, Munch M, Andreasen PA, and Gliemann J. Purified alpha 2-macroglobulin receptor/LDL receptor-related protein binds urokinase plasminogen activator inhibitor type-1 complex. Evidence that the alpha 2-macroglobulin receptor mediates cellular degradation of urokinase receptor-bound complexes. J Biol Chem 267: 14543–14546, 1992. Olbricht CJ, Gutjahr E, Cannon JK, Irmler H, and Tisher CC. Effect of low molecular weight proteins and dextran on renal cathepsin B and L activity. Kidney Int 37: 918–926, 1990. Rothgeb M and Gurd R. Physical methods for the study of myoglobin. Methods Enzymol 52: 473–486, 1978. Salacinski PR, McLean C, Sykes JE, Clement-Jones VV, and Lowry PJ. Iodination of proteins, glycoproteins, and peptides using a solid-phase oxidizing agent, 1,3,4,6-tetrachloro-3 alpha,6 alpha-diphenyl glycoluril (Iodogen). Anal Biochem 117: 136–146, 1981. Vanholder R, Sever MS, Erek E, and Lameire N. Acute renal failure related to the crush syndrome: towards an era of seismo-nephrology? Nephrol Dial Transplant 15: 1517–1521, 2000. Xu D, Kozyraki R, Newman TC, and Fyfe JC. Genetic evidence of an accessory activity required specifically for cubilin brush-border expression and intrinsic factor-cobalamin absorption. Blood 94: 3604–3606, 1999. Zager RA. Rhabdomyolysis and myohemoglobinuric acute renal failure. Kidney Int 49: 314–326, 1996.


www.ajprenal.org