Perspective

Integration of Endocytosis and Signal Transduction by Lipoprotein Receptors

See allHide authors and affiliations

Science's STKE  01 Apr 2003:
Vol. 2003, Issue 176, pp. pe12
DOI: 10.1126/stke.2003.176.pe12

Abstract

The members of the low density lipoprotein receptor (LDLR) gene family are cell surface molecules with diverse functions in cellular metabolism. All LDLR family members are endocytic receptors that mediate the uptake of extracellular cargo into the cell; recent research indicates that they also participate directly in signal transduction. Regulated proteolytic release of the intracellular domain of one of these lipoprotein receptors, the LDLR-related protein 1 (LRP1), has been described, along with the possible role of the released domain in transcriptional regulation. A recent study suggests that megalin, a member of the LDLR gene family that mediates the cellular uptake of vitamin D carrier protein, may also modulate vitamin D-related gene transcription through sequestration of a component of the vitamin D receptor transcriptional complex. May et al. discuss this research in the context of the integration of endocytosis and signaling by this receptor family.

The members of the low density lipoprotein receptor (LDLR) gene family are cell surface molecules with diverse functions in cellular metabolism. In mammals, the LDLR family consists of seven core members (Table 1). All are endocytic receptors that mediate the uptake of extracellular cargo into the cell. Their ligands are diverse and comprise not only lipoproteins (hence the name of the family) but also proteases, complexes of proteases and inhibitors, complexes of vitamins with carrier proteins, and extracellular signaling molecules [reviewed in (1)]. Although these receptors were initially believed to function exclusively as transporters, recent research has revealed that they also participate directly in signal transduction processes (Fig. 1), such as the modulation of ion currents in excitable cells (2) and the regulation of cellular protein kinase activity in response to the binding of extracellular signaling molecules (3-5). Regulated proteolytic release of the intracellular domain of one of these lipoprotein receptors, the LDLR-related protein 1 (LRP1), has also been described (6)--along with the possible role of the released domain in the transcriptional regulation. A new study on the function of megalin, a member of the LDLR gene family that mediates the cellular uptake of vitamin D carrier protein (7), now provides another novel angle on the regulation of signal transduction processes by lipoprotein receptors. Petersen and colleagues identified a novel intracellular interaction partner for megalin in yeast two-hybrid screens. Intriguingly, this newly identified megalin binding protein (MegBP) interacts not only with the megalin cytoplasmic tail, but also with SKIP (SKI-interacting protein), a component of the vitamin D receptor transcriptional regulatory complex (8). This suggests that megalin not only transports vitamin D into the cell, but may also modulate vitamin D-related gene transcription directly through MegBP and SKIP. It is tempting to speculate that megalin might exert its regulatory function on the vitamin D receptor transcriptional complex in a coordinated manner together with the uptake of vitamin D. To our knowledge, this would provide the first example for a mechanism in which cellular uptake of a ligand and direct signaling by the same lipoprotein receptor are closely intertwined.

Fig. 1.

Potential mechanisms of megalin signaling. Megalin endocytoses complexes of vitamin D (VitD) and vitamin D binding protein. The intracellular adaptor molecule MegBP interacts with the proximal part of the megalin intracellular domain, presumably at the proline-rich motif. MegBP also binds the transcriptional regulator SKIP, which is part of the vitamin D receptor (VDR) transcription factor complex. Dependent on or independent of the internalization of megalin and its ligand, either posttranslational modification of MegBP or proteolytic release of the megalin intracellular domain might regulate the subcellular localization of SKIP. Megalin also binds the signaling molecule Sonic hedgehog (Shh) on its extracellular domain. How this interaction influences Shh signal transduction is not known. It is conceivable that transcytosis of Shh is required for its action on specific target tissues. Alternatively, megalin could internalize a complex of Shh and Patched, thereby releasing Patched-mediated inhibition of the Shh signaling receptor Smoothened. Megalin binds various other intracellular adaptor molecules with roles in protein kinase signaling and protein trafficking, including MAGI-1, which contains a nuclear localization sequence. Letters on megalin's cytoplasmic tail represent amino acids in various protein-binding domains. CaMKII, calmodulin-dependent kinase II; Dab2, Disabled-2; EB-1, E2a-Pbx1 activated protein; EGF, epidermal growth factor; Glut1-BP, glucose transporter 1 binding protein; JIP, c-Jun N-terminal kinase interacting protein; MAGI-1, membrane-associated guanylate kinase with inverted orientation-1; MAPK, mitogen-activated protein kinase; MegBP, megalin-binding protein; RxR, retinoid x receptor; SKIP, SKI interacting protein.

Table 1.

Mammalian members of the LDLR family.

Signal Transduction Mechanisms of Lipoprotein Receptors

Ligand uptake and transmembrane signal transduction have generally been regarded as separate and independent processes. Regulatory functions of lipoprotein receptors initially appeared limited to cellular feedback mechanisms that occurred in response to the uptake and accumulation of the transported cargo. The regulatory response to the increase in cellular cholesterol content after the endocytosis of cholesterol-rich low density lipoprotein particles by the LDLR is the best-studied example of this [reviewed in (9)].

The discovery of several cellular scaffolding and adaptor proteins that bind to the cytoplasmic tails of lipoprotein receptors has changed this view (10, 11). Adaptor proteins frequently contain multiple structural domains that allow them to bind to other proteins and assist in the assembly of the multiprotein complexes that are critical for specification of most signal transduction processes. The direct interaction of lipoprotein receptors with intracellular adaptors suggested that these receptors might also be directly involved in the regulation of signaling events. The finding that, indeed, two members of the LDLR gene family, the very low density lipoprotein receptor (VLDLR) and the apolipoprotein receptor 2 (ApoER2), directly regulate neuronal migration during embryogenesis by transmitting a signal across the neuronal membrane confirmed this concept. Both receptors bind the secreted signaling molecule reelin on their extracellular domains. This interaction results in tyrosine phosphorylation of the adaptor molecule disabled-1 (Dab1), which binds to NPxY (asparagine-proline-any-tyrosine) motifs in the cytoplasmic tails of the VLDLR and ApoER2 receptors. Src family kinases are necessary for this phosphorylation, and downstream signaling events include the activation of phosphoinositide 3-kinase and protein kinase B (also known as Akt) with subsequent inhibition of glycogen synthase kinase 3β [reviewed in (12)].

Other lipoprotein receptors also interact directly with extracellular signaling molecules. Binding of platelet-derived growth factor (PDGF) BB to the extracellular domain of the LDLR-related protein 1 (LRP1) has been reported (13). PDGF-dependent tyrosine phosphorylation of LRP1 within the second NPxY motif in the LRP1 cytoplasmic domain, which requires the PDGF receptor β and involves Src-family tyrosine kinases (13, 14), allows the LRP1 receptor to bind Shc and possibly other adaptor proteins that connect LRP1 to intracellular signaling pathways. In another study, the extracellular domain of megalin was found to bind the extracellular signal protein Sonic hedgehog (Shh) (15). The precise consequences of this interaction for Shh signaling are not known. However, it is intriguing that megalin-deficient mice present with developmental anomalies that are very similar to those caused by the lack of shh, suggesting a possible role for megalin in Shh signaling [reviewed in (16)].

Another possible signaling mechanism has recently been described for LRP1. This receptor undergoes a series of proteolytic cleavage steps leading to the release of its intracellular domain from the plasma membrane (6). The subsequent fate and subcellular localization of the released fragment are currently unknown. However, the released cytosolic domains of several other transmembrane proteins translocate to the nucleus, where they can participate in transcriptional regulation. So far, this signaling mechanism has been studied in the greatest detail for the Notch family of transmembrane receptors, which are important not only during embryonic development, but also later. For instance, abnormalities in the vascular wall leading to a form of multi-infarct dementia termed CADASIL (cerebral autosomal dominant angiopathy with subcortical infarcts) are caused by mutations in Notch-3 [reviewed in (17)].

Megalin as a Signaling Receptor

The interaction of the cytoplasmic protein MegBP with the megalin tail could conceivably create a direct link from megalin at the cell surface to the transcriptional machinery in the nucleus. The question thus arises whether megalin, like LRP1, might undergo proteolytic processing and whether release of the cytoplasmic domain might regulate gene transcription through factors such as MegBP. The number of membrane proteins that have been shown to release their cytoplasmic tails is growing rapidly [reviewed in (18)], and one has indeed to wonder whether this is a much more widespread signaling mechanism than previously thought. The enzymatic activity of the intramembranous protease γ-secretase, which catalyzes the proteolytic release of the intracellular domains of these membrane proteins, appears to recognize its substrates only after the bulk of their extracellular domains has been shed, leaving a short extracellular stalk behind (19). There is evidence that shedding of the megalin extracellular domain occurs (20), making megalin a likely substrate for γ-secretase and thus a good candidate for a receptor that utlizes this signaling mechanism.

An alternative possibility for a physiological regulatory function of a megalin, MegBP, and SKIP interaction could involve the regulated sequestration of SKIP away from the nucleus by MegBP and megalin. MegBP contains consensus sites for posttranslational modification by several cytosolic enzymes, including protein kinase C and casein kinase II (8). Phosphorylation of MegBP, possibly in response to ligand binding to megalin or after endocytosis of the receptor, may alter its affinity for SKIP, resulting in either release or binding (and thus sequestration). Whatever the exact mechanism, the spatial distribution of MegBP within the cell is likely to be of functional importance, because overexpression of MegBP in megalin-containing cells, but not in cells that did not express megalin, induced rapid cell death independent of megalin endocytosis (8).

The Role of Other Adaptor Proteins

MegBP is not the only protein that Petersen and colleagues identified as interacting with the megalin cytoplasmic domain. Others include the c-Jun N-terminal kinase (JNK) interacting proteins 1 and 2 (JIP-1 and JIP-2), which were also found independently in earlier screens (10). The physiological role of this interaction has not yet been determined. It is noteworthy, however, that the JIPs do not bind exclusively to megalin, but also to the cytoplasmic tails of LRP1 and ApoER2. The biological significance of this interaction is emphasized by the existence of splice variants of the ApoER2 intracellular domain that differ in their ability to bind JIP-1 and -2 (21). Thus, the regulated interaction of lipoprotein receptors with cellular kinases of the mitogen-activated protein kinase family--in other words, the JNKs--points to a common functional role.

Other adaptor molecules that interact with megalin include disabled-2 (Dab2) (22) and membrane-associated guanylate kinase with inverted orientation 1 (MAGI-1) (23), as well as two additonal interaction partners identified by Petersen et al., glucose transporter 1 binding protein (Glut1BP) and E2a-Pbx1 activated protein (EB1). These proteins are all involved in membrane protein trafficking and receptor clustering. There are two ways in which they could exert their biological functions through megalin. They could directly regulate the localization and endocytosis of megalin or, alternatively, they could serve as scaffolds that recruit megalin into a complex with other membrane proteins, thereby modulating their functions.

Dab2 is a widely expressed adaptor molecule that seems to utilize both mechanisms. Conditional inactivation of the dab2 gene selectively in the mouse embryo, without altering its expression in extraembryonic tissues, results in defects in renal tubular function and urinary loss of vitamin carriers and other proteins, reminiscent of the phenotype of megalin knockout mice. Furthermore, the number of clathrin-coated pits is reduced in the cells of the renal proximal tubuli of these conditional knockout mice. In light of the known binding of Dab2 to the megalin cytoplasmic tail and the interaction of Dab2 with the clathrin adaptor protein AP-2 adaptin, these findings suggest that Dab2 is needed for the efficient endocytosis of megalin and its ligands (24).

If dab2 is deleted unconditionally, embryonic development of the affected mice arrests prior to gastrulation, owing to the lack of Dab2 in certain extraembryonic tissues such as the visceral endoderm. Phenotypically, the resulting abnormalities are reminiscent of those caused by the lack of extracellular signaling molecules of the transforming growth factor-β (TGF-β) family (24). Intriguingly, an RNA chip analysis of megalin-deficient kidneys revealed changes in the transcription of several TGF-β target genes, one of which encodes a transcriptional repressor (TGF-β-stimulated clone 22) (25). A homologous protein, encoded by TGF-β-stimulated clone 22 homologous gene 1, interacts with MegBP. Thus, it seems likely that megalin and Dab2 cooperate in the modulation of TGF-β-dependent signaling. However, megalin does not appear to be essential for Dab2-dependent signaling in the visceral endoderm, because megalin-deficient mice proceed normally through gastrulation (26). This would imply that other Dab2 interacting proteins, possibly other members of the LDL receptor gene family, can compensate for the lack of megalin in this particular step of development.

The interaction of the adaptor molecule EB-1 with the megalin tail provides further evidence for an integration of endocytosis and signaling by lipoprotein receptors. The key to determining how EB-1 affects megalin function may lie in the similarity of EB-1 to another adaptor molecule, Numb (27). Numb is involved in the control of asymmetric cell division and has been shown to negatively regulate signaling through cell surface receptors of the Notch family by a mechanism that might involve the endocytic uptake of Notch receptors (28). Whether EB-1 plays a similar role in the regulation of megalin, or another membrane molecule with which megalin interacts, is currently unknown.

The megalin adaptor protein MAGI-1 is another promising subject for further detailed investigations. It belongs to a family of membrane-associated proteins that participate in the assembly of multiprotein complexes. In addition to other structural motifs, it contains a nuclear localization signal (29), strengthening the case for a possible role of megalin in the regulation of gene transcription in the nucleus. Whether the proteolytic release of the megalin cytoplasmic tail or another posttranslational modification of megalin regulates the subcellular localization of MAGI-1, and what consequences this would have for gene transcription and cellular metabolism, remain to be addressed.

Conclusions and Outlook

Our understanding of lipoprotein receptors as signaling molecules has broadened substantially in the recent past. It is now generally recognized that direct interaction with extracellular signaling molecules and intracellular adaptor proteins is an intrinsic function of this ancient family of multifunctional receptors. For some lipoprotein receptors, the specific signaling pathways and their physiological targets have already been elucidated in considerable detail. Still, a myriad of tantalizing protein interactions awaits further functional characterization. Future investigations should also address the fundamental question of to what extent and in which way endocytosis, on the one hand, and activation of intracellular signaling pathways by lipoprotein receptors, on the other, form a functional unit. The newly described interaction of megalin with an adaptor protein that physically links this receptor for a vitamin D carrier protein to a component of the complex that regulates vitamin D receptor-dependent transcription provides a first glimpse of things to come.

Another important area for future studies concerns the role of lipoprotein receptors in the regulation of the subcellular localization of transcriptional modulators. This includes the related questions of whether other lipoprotein receptors besides LRP1 undergo proteolytic processing leading to the release of their cytoplasmic domains, and whether posttranslational modification of the receptor tails directly or indirectly alters the function of these receptor fragments in the nucleus or other sites of action within the cell. Eventually, this research should allow us to understand how the binding of a growing and bewildering spectrum of ligands by lipoprotein receptors of the LDLR family can transduce and modulate a wide and diverse range of extracellular signals in the target cell.

References

  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
  11. 11.
  12. 12.
  13. 13.
  14. 14.
  15. 15.
  16. 16.
  17. 17.
  18. 18.
  19. 19.
  20. 20.
  21. 21.
  22. 22.
  23. 23.
  24. 24.
  25. 25.
  26. 26.
  27. 27.
  28. 28.
  29. 29.
View Abstract

Navigate This Article