Protein sorting by tyrosine-based signals: adapting to the Ys and wherefores.

The endocytic and secretory pathways of eukaryotic cells consist of an array of membrane-bound compartments, each of which contains a characteristic cohort of transmembrane proteins. Understanding how these proteins are targeted to and maintained within their appropriate compartments will be crucial for unravelling the mysteries of organelle biogenesis and function. A common event in the sorting of many transmembrane proteins is the interaction between a sorting signal in the cytosolic domain of the targeted protein and a component of an organellar protein coat. Here, we summarize recent findings on the mechanism of sorting by one type of signal, characterized by the presence of a critical tyrosine (Y) residue, and attempt to integrate these findings into a hypothetical model for protein sorting in the endocytic and late (post-Golgi) secretory pathways.

Invariant chain trimers are sequestered in the rough endoplasmic reticulum in the absence of association with HLA class II antigens.

HLA class II antigens are heterodimeric cell surface glycoproteins that interact with antigenic peptides to form complexes recognizable by CD4-positive T cells. During their biosynthesis, class II antigens are retained in a post-Golgi compartment in association with the invariant chain, which dissociates before class II cell surface expression. To address whether the invariant chain mediates this post-Golgi retention, its transport and assembly were examined in cells that do not express HLA class II antigens. Pulse-chase analysis and endoglycosidase digestions showed that very little invariant chain proceeded as far as the trans-Golgi in class II-negative cell lines. Immunofluorescence studies suggested that in these cells the invariant chain is sequestered in the RER. Gel filtration and cross-linking data showed that RER-localized invariant chain is present as trimers or aggregated trimers. Multimerization is mediated by lumenal interactions; a proteolytic fragment of the invariant chain corresponding to the lumenal domain remained trimeric as determined by cross-linking analysis. Similar transport and structural characteristics were observed for a pool of excess invariant chain in class II-positive cells, suggesting that an excess of invariant chain in the ER may be important for class II antigen function. These results have important implications for the transport of cellular proteins in general and for the role of the invariant chain in class II antigen biosynthesis.

Protein targeting by tyrosine- and di-leucine-based signals: evidence for distinct saturable components.

Targeting of transmembrane proteins to lysosomes, endosomal compartments, or the trans-Golgi network is largely dependent upon cytoplasmically exposed sorting signals. Among the most widely used signals are those that conform to the tyrosine-based motif, YXXO (where Y is tyrosine, X is any amino acid, and O is an amino acid with a bulky hydrophobic group), and to the di-leucine (or LL) motif. Signals conforming to both motifs have been implicated in protein localization to similar post-Golgi compartments. We have exploited the saturability of sorting to ask whether different YXXO or LL signals use shared components of the targeting machinery. Chimeric proteins containing various cytoplasmic domains and/or targeting signals were overexpressed in HeLa cells by transient transfection. Endogenous transferrin receptor and lysosomal proteins accumulated at the cell surface upon overexpression of chimeric proteins containing functional YXXO targeting signals, regardless of the compartmental destination imparted by the signal. Furthermore, overexpression of these chimeric proteins compromised YXXO-mediated endocytosis and lysosomal delivery. These activities were ablated by mutating the signals or by appending sequences that conformed to the YXXO motif but lacked targeting activity. Interestingly, overexpression of chimeric proteins containing cytoplasmic LL signals failed to induce surface displacement of endogenous YXXO-containing proteins, but did displace other proteins containing LL motifs. Our data demonstrate that: (a) Protein targeting and internalization mediated by either YXXO or LL motifs are saturable processes; (b) common saturable components are used in YXXO-mediated protein internalization and targeting to different post-Golgi compartments; and (c) YXXO- and LL-mediated targeting mechanisms use distinct saturable components.

A dominant-negative clathrin mutant differentially affects trafficking of molecules with distinct sorting motifs in the class II major histocompatibility complex (MHC) pathway.

The role of clathrin in intracellular sorting was investigated by expression of a dominant-negative mutant form of clathrin, termed the hub fragment. Hub inhibition of clathrin-mediated membrane transport was established by demonstrating a block of transferrin internalization and an alteration in the intracellular distribution of the cation-independent mannose-6-phosphate receptor. Hubs had no effect on uptake of FITC-dextran, adaptor distribution, organelle integrity in the secretory pathway, or cell surface expression of constitutively secreted molecules. Hub expression blocked lysosomal delivery of chimeric molecules containing either the tyrosine-based sorting signal of H2M or the dileucine-based sorting signal of CD3gamma, confirming a role for clathrin-coated vesicles (CCVs) in recognizing these signals and sorting them to the endocytic pathway. Hub expression was then used to probe the role of CCVs in targeting native molecules bearing these sorting signals in the context of HLA-DM and the invariant chain (I chain) complexed to HLA-DR. The distribution of these molecules was differentially affected. Accumulation of hubs before expression of the DM dimer blocked DM export from the TGN, whereas hubs had no effect on direct targeting of the DR-I chain complex from the TGN to the endocytic pathway. However, concurrent expression of hubs, such that hubs were building to inhibitory concentrations during DM or DR-I chain expression, caused cell surface accumulation of both complexes. These observations suggest that both DM and DR-I chain are directly transported to the endocytic pathway from the TGN, DM in CCVs, and DR-I chain independent of CCVs. Subsequently, both complexes can appear at the cell surface from where they are both internalized by CCVs. Differential packaging in CCVs in the TGN, mediated by tyrosine- and dileucine-based sorting signals, could be a mechanism for functional segregation of DM from DR-I chain until their intended rendezvous in late endocytic compartments.

Distinct protein sorting and localization to premelanosomes, melanosomes, and lysosomes in pigmented melanocytic cells.

Melanosomes and premelanosomes are lysosome-related organelles with a unique structure and cohort of resident proteins. We have positioned these organelles relative to endosomes and lysosomes in pigmented melanoma cells and melanocytes. Melanosome resident proteins Pmel17 and TRP1 localized to separate vesicular structures that were distinct from those enriched in lysosomal proteins. In immunogold-labeled ultrathin cryosections, Pmel17 was most enriched along the intralumenal striations of premelanosomes. Increased pigmentation was accompanied by a decrease in Pmel17 and by an increase in TRP1 in the limiting membrane. Both proteins were largely excluded from lysosomal compartments enriched in LAMP1 and cathepsin D. By kinetic analysis of fluid phase uptake and immunogold labeling, premelanosomal proteins segregated from endocytic markers within an unusual endosomal compartment. This compartment contained Pmel17, was accessed by BSA-gold after 15 min, was acidic, and displayed a cytoplasmic planar coat that contained clathrin. Our results indicate that premelanosomes and melanosomes represent a distinct lineage of organelles, separable from conventional endosomes and lysosomes within pigmented cells. Furthermore, they implicate an unusual clathrin-coated endosomal compartment as a site from which proteins destined for premelanosomes and lysosomes are sorted.

A lysosomal targeting signal in the cytoplasmic tail of the beta chain directs HLA-DM to MHC class II compartments.

In human B cells, class II molecules of the major histocompatibility complex (MHC-II) accumulate in an endosomal/lysosomal compartment, the MIIC, in which they may encounter and bind peptides. An additional molecule required for MHC-II peptide binding, HLA-DM (DM), has also been localized to the MIIC. Neither the relationship of the MIIC to the endosomal system nor the mechanisms by which DM localizes to the MIIC are understood. To address these issues, DM localization was analyzed in cells that do or do not express MHC-II. DM alpha beta heterodimers were localized in transfected MHC-II-negative HeLa and NRK cells, in the absence of the MHC-II-associated invariant chain, to a prelysosomal/lysosomal compartment by immunofluorescence microscopy. To identify a potential targeting determinant, we analyzed the localization of a chimeric protein, T-T-Mb, in which the cytoplasmic tail of murine DM beta (Mb) was appended to the lumenal and transmembrane domains of a cell surface protein, Tac. Like intact DM, T-T-Mb was localized to a lysosomal compartment in HeLa and NRK cells, as judged by immunofluorescence and immunoelectron microscopy. T-T-Mb was rapidly degraded in this compartment by a process that was blocked by inhibitors of lysosomal proteolysis. The DM beta cytoplasmic tail also mediated internalization of anti-Tac antibody from the cell surface and delivery to lysosomes. Deletion from the DM beta cytoplasmic tail of the tyrosine-based motif, YTPL, resulted in cell surface expression of T-T-Mb and a loss of both degradation and internalization; alanine scanning mutagenesis showed that the Y and L residues were critical for these functions. Similarly, mutation of the same Y residue within full-length DM beta resulted in cell surface expression of DM alpha beta heterodimers. Lastly, T-T-Mb was localized by immunoelectron microscopy to the MIIC in a human B lymphoblastoid cell line. Our results suggest that a motif, YTPL, in the cytoplasmic tail of the beta chain of DM is sufficient for targeting either to lysosomes or to the MIIC.

Dominant-negative inhibition of receptor-mediated endocytosis by a dynamin-1 mutant with a defective pleckstrin homology domain.

The dynamins are 100 kDa GTPases involved in the scission of endocytic vesicles from the plasma membrane [1]. Dynamin-1 is present in solution as a tetramer [2], and undergoes further self-assembly following its recruitment to coated pits to form higher-order oligomers that resemble 'collars' around the necks of nascent coated buds [1] [3]. GTP hydrolysis by dynamin in these collars is thought to accompany the 'pinching off' of endocytic vesicles [1] [4]. Dynamin contains a pleckstrin homology (PH) domain that binds phosphoinositides [5] [6], which in turn enhance both the GTPase activity [5] [7] [8] and self-assembly [9] [10] of dynamin. We recently showed that the dynamin PH domain binds phosphoinositides only when it is oligomeric [6]. Here, we demonstrate that interactions between the dynamin PH domain and phosphoinositides are important for dynamin function in vivo. Full-length dynamin-1 containing mutations that abolish phosphoinositide binding by its PH domain was a dominant-negative inhibitor of receptor-mediated endocytosis. Mutated dynamin-1 with both a defective PH domain and impaired GTP binding and hydrolysis also inhibited receptor-mediated endocytosis. These findings suggest that the role of the PH domain in dynamin function differs from that seen for other PH domains. We propose that high-avidity binding to phosphoinositide-rich regions of the membrane by the multiple PH domains in a dynamin oligomer is critical for dynamin's ability to complete vesicle budding.

Interferon consensus sequence-binding protein, a member of the interferon regulatory factor family, suppresses interferon-induced gene transcription.

We previously isolated a cDNA clone encoding interferon consensus sequence-binding protein (ICSBP), a member of the interferon regulatory factor (IRF) family, that binds to the interferon (IFN)-stimulated response element (ISRE) of many IFN-regulated genes. In this investigation, we studied the functional role of ICSBP by transient cotransfection of ICSBP cDNA with IFN-responsive reporter genes into the human embryonal carcinoma cell line N-Tera2. These cells were shown not to express ICSBP or IRF-2, thus allowing functional analysis of transfected cDNAs. Cotransfection of ICSBP into cells treated with retinoic acid or any of the IFNs (alpha, beta, or gamma) repressed expression of a chloramphenicol acetyltransferase reporter driven by the major histocompatibility complex class I gene promoter. Similarly, ICSBP repressed expression of chloramphenicol acetyltransferase reporters driven by the ISREs of the 2'-5' oligoadenylate synthetase, guanylate-binding protein, and ISG-15 genes in IFN-treated cells. The repression was dependent on the presence of the ISRE in the reporter. Deletion analysis showed that the putative N-terminal DNA binding domain of ICSBP by itself is capable of mediating the repression. Using the same cotransfection conditions as for ICSBP, a similar repression of these reporters was observed with IRF-2. Finally, ICSBP repressed the IRF-1-mediated induction of major histocompatibility complex class I and IFN-beta reporters in the absence of IFN or retinoic acid. Taken together, these results suggest that ICSBP is a negative regulatory factor capable of repressing transcription of target genes induced by IFN, retinoic acid, or IRF-1.

Dominant negative retinoid X receptor beta inhibits retinoic acid-responsive gene regulation in embryonal carcinoma cells.

Retinoid X receptors (RXRs) heterodimerize with multiple nuclear hormone receptors and are thought to exert pleiotropic functions. To address the role of RXRs in retinoic acid- (RA) mediated gene regulation, we designed a dominant negative RXR beta. This mutated receptor, termed DBD-, lacked the DNA binding domain but retained the ability to dimerize with partner receptors, resulting in formation of nonfunctional dimers. DBD- was transfected into P19 murine embryonal carcinoma (EC) cells, in which reporters containing the RA-responsive elements (RAREs) were activated by RA through the activity of endogenous RXR-RA receptor (RAR) heterodimers. We found that DBD- had a dominant negative activity on the RARE reporter activity in these cells. P19 clones stably expressing DBD- were established; these clones also failed to activate RARE-driven reporters in response to RA. Further, these cells were defective in RA-induced mRNA expression of Hox-1.3 and RAR beta, as well as in RA-induced down-regulation of Oct3 mRNA. Gel mobility shift assays demonstrated that RA treatment of control P19 cells induces RARE-binding activity, of which RXR beta is a major component. However, the RA-induced binding activity was greatly reduced in cells expressing DBD-. By genomic footprinting, we show that RA treatment induces in vivo occupancy of the RARE in the endogenous RAR beta gene in control P19 cells but that this occupancy is not observed with the DBD- cells. These data provide evidence that the dominant negative activity of DBD- is caused by the lack of receptor binding to target DNA. Finally, we show that in F9 EC cells expression of DBD- leads to inhibition of the growth arrest that accompanies RA-induced differentiation. Taken together, these results demonstrate that RXR beta and partner receptors play a central role in RA-mediated gene regulation and in the control of growth and differentiation in EC cells.

Inhibition of estrogen-responsive gene activation by the retinoid X receptor beta: evidence for multiple inhibitory pathways.

The retinoid X receptor beta (RXR beta; H-2RIIBP) forms heterodimers with various nuclear hormone receptors and binds multiple hormone response elements, including the estrogen response element (ERE). In this report, we show that endogenous RXR beta contributes to ERE binding activity in nuclear extracts of the human breast cancer cell line MCF-7. To define a possible regulatory role of RXR beta regarding estrogen-responsive transcription in breast cancer cells, RXR beta and a reporter gene driven by the vitellogenin A2 ERE were transfected into estrogen-treated MCF-7 cells. RXR beta inhibited ERE-driven reporter activity in a dose-dependent and element-specific fashion. This inhibition occurred in the absence of the RXR ligand 9-cis retinoic acid. The RXR beta-induced inhibition was specific for estrogen receptor (ER)-mediated ERE activation because inhibition was observed in ER-negative MDA-MB-231 cells only following transfection of the estrogen-activated ER. No inhibition of the basal reporter activity was observed. The inhibition was not caused by simple competition of RXR beta with the ER for ERE binding, since deletion mutants retaining DNA binding activity but lacking the N-terminal or C-terminal domain failed to inhibit reporter activity. In addition, cross-linking studies indicated the presence of an auxiliary nuclear factor present in MCF-7 cells that contributed to RXR beta binding of the ERE. Studies using known heterodimerization partners of RXR beta confirmed that RXR beta/triiodothyronine receptor alpha heterodimers avidly bind the ERE but revealed the existence of another triiodothyronine-independent pathway of ERE inhibition. These results indicate that estrogen-responsive genes may be negatively regulated by RXR beta through two distinct pathways.

Endocytic clathrin-coated pit formation is independent of receptor internalization signal levels.

The mechanisms responsible for coated pit formation in cells remain unknown, but indirect evidence has argued both for and against a critical role of receptor cytoplasmic domains in the process. If the endocytic motifs of receptors are responsible for recruiting AP2 to the plasma membrane, thereby driving coated pit formation, then the level of constitutively internalized receptors at the membrane would be expected to govern the steady-state level of coated pits in cells. Here we directly test this hypothesis for broad classes of receptors containing three distinct constitutive internalization signals. Chimeric proteins consisting of an integral membrane reporter protein (Tac) coupled to cytoplasmic domains bearing tyrosine-, di-leucine-, or acidic cluster/casein kinase II-based internalization signals were overexpressed to levels that saturated the internalization pathway. Quantitative confocal immunofluorescence microscopy indicated that the number of plasma membrane clathrin-coated pits and the concentration of their structural components were invariant when comparing cells expressing saturating levels of the chimeric receptors to nonexpressing cells or to cells expressing only the Tac reporter lacking cytoplasmic internalization signals. Biochemical analysis showed that the distribution of coat proteins between assembled coated pits and soluble pools was also not altered by receptor overexpression. Finally, the cellular localizations of AP2 and AP1 were similarly unaffected. These results provide a clear indication that receptor endocytic signals do not determine coated pit levels by directly recruiting AP2 molecules. Rather, the findings support a model in which coated pit formation proceeds through recruitment and activation of AP2, likely through a limited number of regulated docking sites that act independently of endocytic signals.

Pmel17 initiates premelanosome morphogenesis within multivesicular bodies.

Melanosomes are tissue-specific organelles within which melanin is synthesized and stored. The melanocyte-specific glycoprotein Pmel17 is enriched in the lumen of premelanosomes, where it associates with characteristic striations of unknown composition upon which melanin is deposited. However, Pmel17 is synthesized as an integral membrane protein. To clarify its physical linkage to premelanosomes, we analyzed the posttranslational processing of human Pmel17 in pigmented and transfected nonpigmented cells. We show that Pmel17 is cleaved in a post-Golgi compartment into two disulfide-linked subunits: a large lumenal subunit, M alpha, and an integral membrane subunit, M beta. The two subunits remain associated intracellularly, indicating that detectable M alpha remains membrane bound. We have previously shown that Pmel17 accumulates on intralumenal membrane vesicles and striations of premelanosomes in pigmented cells. In transfected nonpigmented cells Pmel17 associates with the intralumenal membrane vesicles of multivesicular bodies; cells overexpressing Pmel17 also display structures resembling premelanosomal striations within these compartments. These results suggest that Pmel17 is sufficient to drive the formation of striations from within multivesicular bodies and is thus directly involved in the biogenesis of premelanosomes.

Biosynthesis and processing of class II histocompatibility antigens.

The class II major histocompatibility antigens at the cell surface exist as heterodimers of alpha and beta subunits. During biosynthesis, these subunits are associated with a third chain, the invariant (I) or I chain. Association with the I chain occurs early in biosynthesis in the rough endoplasmic reticulum and persists during transport through the Golgi apparatus. One of the two alpha subunit N-linked oligosaccharides and the single beta subunit N-linked oligosaccharide are converted to the complex form during Golgi transit. In the human system, both I chain N-linked oligosaccharides can also be processed to the complex form, and at least two O-linked oligosaccharides can be added to the I chain. At some point during transit to the cell surface, class II antigens associate with a proteoglycan bearing chondroitin sulfate side chains. Complexes containing alpha, beta and I chain subunits and the associated proteoglycan accumulate in human B-cell lines treated with the ionophore monensin, an inhibitor of Golgi transport, suggesting that this may be a biosynthetic intermediate in class II antigen transport and assembly. Prior to cell surface expression of class II antigens, the exocytic pathway which they follow intersects the endocytic route, followed by certain ligands internalized by receptor-mediated endocytosis. The I chain appears to dissociate from mature class II alpha, beta dimers prior to their cell surface expression but following the intersection of the exocytic and endocytic pathways.

Characterization of the chimeric retinoic acid receptor RARalpha/VDR.

The chimeric receptor, RARalpha/VDR, contains the DNA-binding domain of the retinoic acid receptor (RARalpha) and the ligand-binding domain of the vitamin D receptor (VDR). The ligand-binding properties of RARalpha/VDR are equivalent to that of VDR, with an observed Kd for 1alpha,25 dihydroxy-vitamin D3 (D3) of 0.5 nM. In CV-1 cells, both RARalpha and RARalpha/VDR induce comparable levels of ligand-mediated transcriptional activity from the retinoic acid responsive reporter gene, beta(RARE)3-TK-luciferase, in the presence of the ligand predicted from the receptor ligand-binding domain. Two chimeric RAR receptors were constructed which contained the ligand-binding domain of the estrogen receptor (ER): RARalpha/ER and ER/RARalpha/ER. Both RARalpha/ER and ER/RARalpha/ER bind beta-estradiol with high affinity, and are transcriptionally active only from palindromic RAREs (TREpal and/or (TRE3)3). Only RARalpha/VDR matched in kind and degree the functional characteristics of RARalpha: (1) maximally active from the beta(RARE); (2) moderately active from the TREs; (3) inactive from the retinoic X receptor response elements (RXREs) ApoA1 and CRBP II; (4) forms heterodimers with RXRalpha; and (5) binds to the betaRARE. F9 embryonal carcinoma cell lines were generated which express RARalpha/VDR mRNA (F9RARalpha/VDR cells) and compared with F9 wild-type (F9-Wt) cells, which do not express VDR mRNA. Treatment with all-trans retinoic acid (tRA) inhibits cell growth and induces the differentiation morphology in both F9-Wt and F9-RARalpha/VDR cells; whereas, treatment with D3 is similarly effective only for F9-RARalpha/VDR cells. It is concluded RARalpha/VDR is an useful 'tool' to pinpoint, or to augment transcription from RAREs in gene pathways controlled by RAR without inhibiting the retinoid responsiveness of endogenous RARs.

Heterodimerization among thyroid hormone receptor, retinoic acid receptor, retinoid X receptor, chicken ovalbumin upstream promoter transcription factor, and an endogenous liver protein.

Thyroid hormone receptor (TR) binds to DNA as a monomer, homodimer, and heterodimer with nuclear proteins. We have confirmed that the TR can heterodimerize with retinoid X receptors (RXRs)-alpha and -beta, and have found that another member of the nuclear receptor superfamily, chicken ovalbumin upstream promoter transcription factor (COUP-TF), also formed heterodimers with the TR in the context of binding to a palindromic thyroid hormone-responsive element (TREp). The interaction between COUP-TF and the TR was confirmed using specific antibodies which supershifted the COUP-TF/TR DNA complexes. The complex between the TR and the major TR heterodimerization partner in liver was unaffected by antibodies to COUP-TF and RXR beta, but was supershifted by an anti-RXR alpha antibody, indicating that the liver protein is highly related to RXR alpha. Indeed, the TR/RXR and TR/liver protein heterodimers contact the same guanidine residues in TREp. The retinoic acid receptor (RAR) also heterodimerized with COUP-TF as well as with RXR alpha, RXR beta, and the TR heterodimerization partner in liver. In contrast to its ability to heterodimerize with the TR and RAR, we did not detect heterodimers between COUP-TF and either RXR alpha, RXR beta, or the liver nuclear protein in the context of binding to the TREp. These results show that the major TR heterodimerization partner in liver is highly related to RXR alpha, but that other nuclear receptors such as COUP-TF can heterodimerize with the TR and RAR, suggesting that selective protein-protein interactions may be involved in the tissue and target gene specificities of hormone action.

H-2RIIBP expressed from a baculovirus vector binds to multiple hormone response elements.

H-2RIIBP is a member of the nuclear hormone receptor superfamily that binds to the region II enhancer of major histocompatibility complex class I genes. Based on its homology with Drosophila XR2C/CF1, H-2RIIBP may play a role in development. By using a baculovirus expression system, a large amount of recombinant H-2RIIBP was produced. The recombinant protein accumulated in the nucleus of insect cells. A series of monoclonal antibodies reacting with the recombinant H-2RIIBP was then generated. A DNA-protein immunoprecipitation assay was developed with these antibodies, enabling the DNA-binding specificity of H-2RIIBP to be distinguished from that of an endogenous region II binding factor expressed in uninfected insect cells. We show that H-2RIIBP binds to estrogen response elements with an affinity comparable to that for the region II enhancer. H-2RIIBP also bound to some, but not all, thyroid hormone response elements and retinoic acid response elements, albeit at a lower affinity. Binding to these elements was demonstrated without exogenous addition of a ligand. The H-2RIIBP binding specificity determined by this assay was in agreement with the specificity assessed by Southwestern and gel mobility shift assays. Furthermore, methylation interference assays indicated that H-2RIIBP recognizes the conserved hormone response motif GG(T/A)CA. Taken together, these data demonstrate that H-2RIIBP is capable of binding to hormone response elements of a variety of genes. They suggest that H-2RIIBP may exert a pleiotropic function.

Intramedullary megakaryocytes internalize released platelet factor 4 and store it in alpha granules.

BACKGROUND: Megakaryocytes express and store platelet factor 4 (PF4) in alpha granules. In vivo, PF4 is a clinically relevant, negative regulator of megakaryopoiesis and hematopoietic stem cell replication. These findings would suggest a regulated source of free intramedullary PF4. OBJECTIVES: Define the source of free intramedullary PF4 and its intramedullary life cycle. METHODS: We interrogated both murine and human bone marrow-derived cells during megakaryopoiesis in vitro by using confocal microscopy and enzyme-linked immunosorbent assay. With immunohistochemistry, we examined in vivo free PF4 in murine bone marrow before and after radiation injury and in the setting of megakaryocytopenia and thrombocytopenia. RESULTS: Exogenously added human PF4 is internalized by murine megakaryocytes. Human megakaryocytes similarly take up murine PF4 but not the related chemokine, platelet basic protein. Confocal microscopy shows that internalized PF4 colocalizes with endogenous PF4 in alpha granules and is available for release on thrombin stimulation. Immunohistochemistry shows free PF4 in the marrow, but not another alphagranule protein, von Willebrand factor. Free PF4 increases with radiation injury and decreases with megakaryocytopenia. Consistent with the known role of low-density lipoprotein receptor-related protein 1 in the negative paracrine effect of PF4 on megakaryopoiesis, PF4 internalization is at least partially low-density lipoprotein receptor-related protein 1 dependent. CONCLUSIONS: PF4 has a complex intramedullary life cycle with important implications in megakaryopoiesis and hematopoietic stem cell replication not seen with other tested alpha granule proteins.

Protein sorting within the MHC class II antigen-processing pathway.

Major histocompatibility complex (MHC) class II molecules are required for the presentation of antigenic peptides that are derived predominantly from internalized proteins. The assembly of MHC class II/peptide complexes occurs within endosomal compartments of antigen-presenting cells (APCs). Therefore, for assembly to occur, MHC class II molecules, foreign proteins, and accessory molecules must be sorted to appropriate intracellular sites. My laboratory is trying to understand how proteins are sorted to various antigen-processing compartments as well as to conventional endosomal organelles. Using chimeric marker proteins and a variety of biochemical and genetic approaches, we are addressing the specificity of protein sorting and the mechanisms by which sorting signals are deciphered. By using a similar chimeric protein approach to target endogenous proteins to distinct compartments, we hope to address the role of processing events in each compartment in the generation of MHC class II ligands.