Lrp1/LDL receptor play critical roles in mannose 6-phosphate-independent lysosomal enzyme targeting
Sandra Markmann1†, Melanie Thelen2†, Kerstin Cornils3, Michaela Schweizer4, Nahal BrockeAhmadinejad2, Thomas Willnow6, Joerg Heeren5, Volkmar Gieselmann2, Thomas Braulke1*, and Katrin Kollmann1 1
Department for Biochemistry, Children’s Hospital, University Medical Center Hamburg-Eppendorf, D-
20246 Hamburg, Germany 2
Institute of Biochemistry and Molecular Biology, University of Bonn, Nussallee 11, D-53115 Bonn,
Germany 3
Research Department Cell and Gene Therapy, Clinic for Stem Cell Transplantation, University
Medical Center Hamburg-Eppendorf, D-20246 Hamburg, Germany. 4
Center for Molecular Neurobiology Hamburg, ZMNH, University Medical Center Hamburg-Eppendorf,
20246 Hamburg, Germany. 5
Department of Biochemistry and Molecular Cell Biology, University Medical Center Hamburg-
Eppendorf, D-20246 Hamburg, Germany 6 †
Max Delbrück Center for Molecular Medicine, 13125 Berlin-Buch, Germany The first two authors contributed equally
*Corresponding author: Thomas Braulke, Department for Biochemistry, Children’s Hospital, University Medical Center Hamburg-Eppendorf , D-20246 Hamburg, Germany, Phone: +49 40 741054493, Fax: + 49 40 741058504, [email protected]
Key words: low density lipoprotein-related receptor, lysosomal storage disorder, lysosomal targeting, mannose 6-phosphate, receptor-mediated endocytosis, sortilin
This article has been accepted for publication and undergone full peer review but has not been through the copyediting, typesetting, pagination and proofreading process, which may lead to differences between this version and the Version of Record. Please cite this article as doi: 10.1111/tra.12284
This article is protected by copyright. All rights reserved
Synopsis SILAC-based quantitative analysis of the lysosomal proteome of mouse fibroblasts allowed the identification of lysosomal enzymes whose localisation was unaffected by the lack of mannose 6phosphate targeting signals. Among these, the sorting of cathepsins D and B is independent of sortilin, but involves secretion-recapture mechanisms mediated by Lrp1 and LDL receptors.
Abstract Most lysosomal enzymes require mannose 6phosphate (M6P) residues for efficient receptormediated lysosomal targeting. Although the lack of M6P residues results in missorting and hypersecretion, selected lysosomal enzymes reach normal levels in lysosomes of various cell types suggesting the existence of M6P-independent transport routes. Here, we quantify the lysosomal proteome in M6P-deficient mouse fibroblasts (PTki) using Stable Isotope Labeling by Amino acids in Cell culture (SILAC)-based comparative mass spectrometry, and find unchanged amounts of 20 % of lysosomal enzymes, including cathepsin D and B (Ctsd, Ctsb). Examination of fibroblasts from a new mouse line lacking both M6P and sortilin, a candidate for M6P-independent transport of lysosomal enzymes, revealed that sortilin does not act as cargo receptor for Ctsb and Ctsd. Using fibroblast lines deficient for endocytic lipoprotein receptors we could demonstrate that both LDL receptor and Lrp1 mediate the internalization of non-phosphorylated Ctsb and Ctsd. Furthermore, the presence of Lrp1 inhibitor increased the secretion of Ctsd from PTki cells. These findings establish Lrp1 and LDL receptors in M6Pindependent secretion-recapture targeting mechanism for lysosomal enzymes. Introduction Newly synthesized soluble lysosomal enzymes require a modification on their N-linked oligosaccharides with mannose 6-phosphate (M6P) residues for their efficient targeting to lysosomes. Two enzymes are involved in the sequential formation of the M6P marker. First, the Nacetylglucosamine (GlcNAc)-1-phosphotransferase (PT), localized in the cis Golgi apparatus, catalyzes the transfer of GlcNAc-1-phosphate from UDPGlcNAc to selected C6 hydroxyl groups of one or more mannoses of high mannose type oligosaccharides of lysosomal enzymes (1, 2). The second enzyme, the GlcNAc-1 phosphodiester α-Nacetylglucosaminidase (uncovering enzyme) is localized in the trans-Golgi network (TGN) and
This article is protected by copyright. All rights reserved
removes the GlcNAc residue to expose M6P (3, 4). M6P-containing enzymes can be recognized by two distinct M6P receptors, MPR46 and MPR300, in late Golgi compartments (5). At least 5-20% of newly synthesized lysosomal proteins escape binding to MPR and are secreted in murine and human fibroblasts, and several cell lines. The packaging of MPR-ligand complexes into clathrincoated transport vesicles in the TGN and their endosomal delivery depend on sorting signals present in the cytoplasmic domains of both MPR (6). Upon arrival, the acidic pH of endosomes induces the dissociation of MPR-ligand complexes and recycling of MPR to the TGN or cell surface. The mechanism by which lumenal lysosomal proteins are transferred from endosomes to lysosomes is a matter of debate (reviewed in (7)). Finally, lysosomal proteins are dephosphorylated and several of them undergo proteolytic activation (8, 9). The PT complex is composed of two α-, two β- and two γ-subunits which are encoded by two genes. GNPTAB encodes a precursor protein of 1256 amino acids containing two transmembrane domains (10, 11) that is proteolytically cleaved by the site-1 protease in the Golgi apparatus to form catalytically active α- and β-subunits (12). The function of the soluble γ-subunit encoded by GNPTG is not defined (13). Mutations in GNPTAB resulting in a total loss of PT activity cause a severe lysosomal storage disorder, mucolipidosis type II (MLII, I-cell disease), that is characterized by progressive neurodegeneration, growth retardation, skeletal deformities (dysostosis multiplex), cardiorespiratory defects and early death between 5 and 8 years of age (14). The failure to form M6P residues leads to missorting and hypersecretion of lysosomal enzymes due to their inability to bind to MPR. The subsequent deficiency of multiple lysosomal hydrolases results in lysosomal dysfunction and accumulation of non-degraded material. However, many cell types such as hepatocytes, Kupffer cells, and leukocytes, as well as several organs such as liver, spleen, kidney and brain of MLII patients have nearly normal level of certain lysosomal enzymes (2, 15, 16). These observations suggest the existence of alternate M6P-independent pathways for lysosomal enzyme sorting.
It has been reported that the pro-neurotrophin receptor sortilin mediates the intracellular targeting of subsets of lysosomal proteins such as acid sphingomyelinase, prosaposin, cathepsin D and H (17-19) using endosomal/ lysosomal sorting motifs with a strong homology to the MPR300 (20). Sortilin is a member of the Vps10 domain receptor family, which was named after a yeast vacuolar sorting protein that mediates the transport of carboxypeptidase Y from the TGN to the vacuole (21). The Vps10 domain is responsible for the binding of neurotensin and the receptor-associated protein (RAP), a chaperone for members of the LDL receptor (LDLR) family (22, 23). To identify lysosomal enzymes transported to lysosomes in the absence of M6P residues, we performed a Stable Isotope Labeling by Amino acids in Cell culture (SILAC)-based comparative mass spectrometric analysis of isolated lysosomes from fibroblasts of phosphotransferase deficient ki knock-in mice (PT ) exhibiting cell symptoms of MLII disease (24). We found normal amounts of selected lysosomal enzymes such as cathepsin D and cathepsin B, in PTki lysosomes, indicating that their targeting is independent of M6P. Using various fibroblast cell lines derived from mice defective in single or combined cargo receptors with or without M6P-targeting signals, we demonstrated that nonphosphorylated cathepsin D and cathepsin B are transported to lysosomes via secretion and recapture mechanisms which involve the LDL receptor, and LDL receptor-related protein 1 (Lrp1) but not sortilin. These findings might have impact to improve efficiency and tissue targeting of M6Pdepending enzyme replacement therapies for other lysosomal diseases. Results Protein-specific loss of soluble lysosomal proteins lacking M6P residues Effective targeting of newly synthesized soluble lysosomal proteins to the lysosomal compartment requires the modification of their N-linked glycans with M6P residues. Several studies, however, have reported M6P-independent transport of distinct lysosomal proteins to lysosomes (15, 25, 26). On the other hand, the analysis of tissues or cultured cells derived from GlcNAc-1-phosphotransferase ki deficient mice (PT ) revealed that distinct lysosomal proteins exhibit different requirements of M6P residues for lysosomal targeting (24). To determine the relative amount of all soluble proteins at steady state in lysosomes, we performed a SILAC-based comparative quantitative analysis of lysosomal fractions isolated from fibroblasts of wild-type (wt) and PTki mice. The PTki and wt control cells were grown in either light or heavy SILAC-media for six passages, respectively (Figure 1A). The cells were then incubated with magnetite-containing culture medium for 24 h followed by a 36 h chase period in magnetite-free heavy or light medium, respectively. Equal amounts of postnuclear supernatant (PNS) were mixed, magnetite-containing lysosomal fractions separated on a magnetic column, and the eluate was processed for MS analysis (Figure 1A). Among up to 2,971 proteins identified quantified (the mass spectrometry dataset has been deposited
This article is protected by copyright. All rights reserved
to the Proteome Xchange Consortium (27) via the PRIDE partner repository with the dataset identifier PXD001221), we found 54 known soluble lysosomal proteins (Table S1) of which 51 were detected in amounts sufficient for quantification. These proteins were plotted against the calculated heavy to light ratio (H/L ratio), displaying their depletion in lysosomes of PTki fibroblasts compared to wt lysosomes (Figure 1B). Although some proteins like CLN5 show a considerable variation, the regulation of all displayed proteins shows a corrected p-value
Sandra Markmann1†, Melanie Thelen2†, Kerstin Cornils3, Michaela Schweizer4, Nahal BrockeAhmadinejad2, Thomas Willnow6, Joerg Heeren5, Volkmar Gieselmann2, Thomas Braulke1*, and Katrin Kollmann1 1
Department for Biochemistry, Children’s Hospital, University Medical Center Hamburg-Eppendorf, D-
20246 Hamburg, Germany 2
Institute of Biochemistry and Molecular Biology, University of Bonn, Nussallee 11, D-53115 Bonn,
Germany 3
Research Department Cell and Gene Therapy, Clinic for Stem Cell Transplantation, University
Medical Center Hamburg-Eppendorf, D-20246 Hamburg, Germany. 4
Center for Molecular Neurobiology Hamburg, ZMNH, University Medical Center Hamburg-Eppendorf,
20246 Hamburg, Germany. 5
Department of Biochemistry and Molecular Cell Biology, University Medical Center Hamburg-
Eppendorf, D-20246 Hamburg, Germany 6 †
Max Delbrück Center for Molecular Medicine, 13125 Berlin-Buch, Germany The first two authors contributed equally
*Corresponding author: Thomas Braulke, Department for Biochemistry, Children’s Hospital, University Medical Center Hamburg-Eppendorf , D-20246 Hamburg, Germany, Phone: +49 40 741054493, Fax: + 49 40 741058504, [email protected]
Key words: low density lipoprotein-related receptor, lysosomal storage disorder, lysosomal targeting, mannose 6-phosphate, receptor-mediated endocytosis, sortilin
This article has been accepted for publication and undergone full peer review but has not been through the copyediting, typesetting, pagination and proofreading process, which may lead to differences between this version and the Version of Record. Please cite this article as doi: 10.1111/tra.12284
This article is protected by copyright. All rights reserved
Synopsis SILAC-based quantitative analysis of the lysosomal proteome of mouse fibroblasts allowed the identification of lysosomal enzymes whose localisation was unaffected by the lack of mannose 6phosphate targeting signals. Among these, the sorting of cathepsins D and B is independent of sortilin, but involves secretion-recapture mechanisms mediated by Lrp1 and LDL receptors.
Abstract Most lysosomal enzymes require mannose 6phosphate (M6P) residues for efficient receptormediated lysosomal targeting. Although the lack of M6P residues results in missorting and hypersecretion, selected lysosomal enzymes reach normal levels in lysosomes of various cell types suggesting the existence of M6P-independent transport routes. Here, we quantify the lysosomal proteome in M6P-deficient mouse fibroblasts (PTki) using Stable Isotope Labeling by Amino acids in Cell culture (SILAC)-based comparative mass spectrometry, and find unchanged amounts of 20 % of lysosomal enzymes, including cathepsin D and B (Ctsd, Ctsb). Examination of fibroblasts from a new mouse line lacking both M6P and sortilin, a candidate for M6P-independent transport of lysosomal enzymes, revealed that sortilin does not act as cargo receptor for Ctsb and Ctsd. Using fibroblast lines deficient for endocytic lipoprotein receptors we could demonstrate that both LDL receptor and Lrp1 mediate the internalization of non-phosphorylated Ctsb and Ctsd. Furthermore, the presence of Lrp1 inhibitor increased the secretion of Ctsd from PTki cells. These findings establish Lrp1 and LDL receptors in M6Pindependent secretion-recapture targeting mechanism for lysosomal enzymes. Introduction Newly synthesized soluble lysosomal enzymes require a modification on their N-linked oligosaccharides with mannose 6-phosphate (M6P) residues for their efficient targeting to lysosomes. Two enzymes are involved in the sequential formation of the M6P marker. First, the Nacetylglucosamine (GlcNAc)-1-phosphotransferase (PT), localized in the cis Golgi apparatus, catalyzes the transfer of GlcNAc-1-phosphate from UDPGlcNAc to selected C6 hydroxyl groups of one or more mannoses of high mannose type oligosaccharides of lysosomal enzymes (1, 2). The second enzyme, the GlcNAc-1 phosphodiester α-Nacetylglucosaminidase (uncovering enzyme) is localized in the trans-Golgi network (TGN) and
This article is protected by copyright. All rights reserved
removes the GlcNAc residue to expose M6P (3, 4). M6P-containing enzymes can be recognized by two distinct M6P receptors, MPR46 and MPR300, in late Golgi compartments (5). At least 5-20% of newly synthesized lysosomal proteins escape binding to MPR and are secreted in murine and human fibroblasts, and several cell lines. The packaging of MPR-ligand complexes into clathrincoated transport vesicles in the TGN and their endosomal delivery depend on sorting signals present in the cytoplasmic domains of both MPR (6). Upon arrival, the acidic pH of endosomes induces the dissociation of MPR-ligand complexes and recycling of MPR to the TGN or cell surface. The mechanism by which lumenal lysosomal proteins are transferred from endosomes to lysosomes is a matter of debate (reviewed in (7)). Finally, lysosomal proteins are dephosphorylated and several of them undergo proteolytic activation (8, 9). The PT complex is composed of two α-, two β- and two γ-subunits which are encoded by two genes. GNPTAB encodes a precursor protein of 1256 amino acids containing two transmembrane domains (10, 11) that is proteolytically cleaved by the site-1 protease in the Golgi apparatus to form catalytically active α- and β-subunits (12). The function of the soluble γ-subunit encoded by GNPTG is not defined (13). Mutations in GNPTAB resulting in a total loss of PT activity cause a severe lysosomal storage disorder, mucolipidosis type II (MLII, I-cell disease), that is characterized by progressive neurodegeneration, growth retardation, skeletal deformities (dysostosis multiplex), cardiorespiratory defects and early death between 5 and 8 years of age (14). The failure to form M6P residues leads to missorting and hypersecretion of lysosomal enzymes due to their inability to bind to MPR. The subsequent deficiency of multiple lysosomal hydrolases results in lysosomal dysfunction and accumulation of non-degraded material. However, many cell types such as hepatocytes, Kupffer cells, and leukocytes, as well as several organs such as liver, spleen, kidney and brain of MLII patients have nearly normal level of certain lysosomal enzymes (2, 15, 16). These observations suggest the existence of alternate M6P-independent pathways for lysosomal enzyme sorting.
It has been reported that the pro-neurotrophin receptor sortilin mediates the intracellular targeting of subsets of lysosomal proteins such as acid sphingomyelinase, prosaposin, cathepsin D and H (17-19) using endosomal/ lysosomal sorting motifs with a strong homology to the MPR300 (20). Sortilin is a member of the Vps10 domain receptor family, which was named after a yeast vacuolar sorting protein that mediates the transport of carboxypeptidase Y from the TGN to the vacuole (21). The Vps10 domain is responsible for the binding of neurotensin and the receptor-associated protein (RAP), a chaperone for members of the LDL receptor (LDLR) family (22, 23). To identify lysosomal enzymes transported to lysosomes in the absence of M6P residues, we performed a Stable Isotope Labeling by Amino acids in Cell culture (SILAC)-based comparative mass spectrometric analysis of isolated lysosomes from fibroblasts of phosphotransferase deficient ki knock-in mice (PT ) exhibiting cell symptoms of MLII disease (24). We found normal amounts of selected lysosomal enzymes such as cathepsin D and cathepsin B, in PTki lysosomes, indicating that their targeting is independent of M6P. Using various fibroblast cell lines derived from mice defective in single or combined cargo receptors with or without M6P-targeting signals, we demonstrated that nonphosphorylated cathepsin D and cathepsin B are transported to lysosomes via secretion and recapture mechanisms which involve the LDL receptor, and LDL receptor-related protein 1 (Lrp1) but not sortilin. These findings might have impact to improve efficiency and tissue targeting of M6Pdepending enzyme replacement therapies for other lysosomal diseases. Results Protein-specific loss of soluble lysosomal proteins lacking M6P residues Effective targeting of newly synthesized soluble lysosomal proteins to the lysosomal compartment requires the modification of their N-linked glycans with M6P residues. Several studies, however, have reported M6P-independent transport of distinct lysosomal proteins to lysosomes (15, 25, 26). On the other hand, the analysis of tissues or cultured cells derived from GlcNAc-1-phosphotransferase ki deficient mice (PT ) revealed that distinct lysosomal proteins exhibit different requirements of M6P residues for lysosomal targeting (24). To determine the relative amount of all soluble proteins at steady state in lysosomes, we performed a SILAC-based comparative quantitative analysis of lysosomal fractions isolated from fibroblasts of wild-type (wt) and PTki mice. The PTki and wt control cells were grown in either light or heavy SILAC-media for six passages, respectively (Figure 1A). The cells were then incubated with magnetite-containing culture medium for 24 h followed by a 36 h chase period in magnetite-free heavy or light medium, respectively. Equal amounts of postnuclear supernatant (PNS) were mixed, magnetite-containing lysosomal fractions separated on a magnetic column, and the eluate was processed for MS analysis (Figure 1A). Among up to 2,971 proteins identified quantified (the mass spectrometry dataset has been deposited
This article is protected by copyright. All rights reserved
to the Proteome Xchange Consortium (27) via the PRIDE partner repository with the dataset identifier PXD001221), we found 54 known soluble lysosomal proteins (Table S1) of which 51 were detected in amounts sufficient for quantification. These proteins were plotted against the calculated heavy to light ratio (H/L ratio), displaying their depletion in lysosomes of PTki fibroblasts compared to wt lysosomes (Figure 1B). Although some proteins like CLN5 show a considerable variation, the regulation of all displayed proteins shows a corrected p-value
