www.elsevier.com/locate/ymcne Mol. Cell. Neurosci. 27 (2004) 370 – 378
Peripheral myelin protein 22 kDa and protein zero: domain specific trans-interactions Birgit Hasse,a,1 Frank Bosse,a,1 Helmut Hanenberg,b and Hans Werner Mqllera,* a
Molecular Neurobiology Laboratory, Department of Neurology, Heinrich-Heine-University, 40225 Dq sseldorf, Germany Department of Pediatric Oncology, Hematology and Immunology, Heinrich-Heine-University, 40225 Dq sseldorf, Germany
b
Received 16 March 2004; revised 18 May 2004; accepted 20 June 2004 Available online 28 September 2004
The peripheral myelin proteins P0 and PMP22 are associated in preparations of compact myelin and in cell cultures coexpressing both molecules. The mechanism of this interaction, however, still needs to be unravelled. We have established three different (cell–cell, cell–protein, protein–protein based) assay systems using retrovirally transduced HeLa cells that overexpressed either PMP22 or P0 and purified GST fusion oligopeptides of PMP22 and P0 to detect domain-specific interactions between these proteins. The results revealed that PMP22 and P0 are involved in both trans-homophilic and trans-heterophilic interactions. Moreover, the data clearly indicate that the heterophilic trans-interaction is mediated through the second loop of PMP22, while the first loop is involved in homophilic trans-interaction of PMP22 proteins. Both modes of interaction are due to direct protein–protein binding. In addition, we demonstrate that disease-related point mutations of P0 resulted in a decreased adhesion capability correlating with the severity of the respective disease phenotype. D 2004 Published by Elsevier Inc.
Introduction The myelin sheath of peripheral nervous system (PNS) is generated by Schwann cells, from which the myelin radiates and wraps around adjacent axons (Arroyo and Scherer, 2000). Peripheral myelination requires the synthesis of large quantities of membrane lipids and a specific set of proteins that play key roles in the structure and function of the myelin sheath. Two important molecules of peripheral myelin are the protein zero (P0), which makes up approximately 50% of all myelin proteins (Greenfield et al., 1973), and the peripheral myelin protein 22
* Corresponding author. Molecular Neurobiology Laboratory, Department of Neurology, Heinrich-Heine-University, Moorenstrasse 5, 40225 Dqsseldorf, Germany. Fax: +49 2118118411. E-mail address:
[email protected] (H.W. Mqller). 1 These authors contributed equally to this study. Available online on ScienceDirect (www.sciencedirect.com.) 1044-7431/$ - see front matter D 2004 Published by Elsevier Inc. doi:10.1016/j.mcn.2004.06.009
kDa (PMP22), which comprises 2–5% of total PNS myelin (Pareek et al., 1993). The genes encoding P0 and PMP22 have been well characterized, and numerous mutations associated with these genes have been identified in hereditary demyelinating peripheral neuropathies in mice and human, such as Charcot–Marie–Tooth disease (CMT), Dejerine–Sottas Syndrome (DSS), Congenital Hypomyelination (CH) and Hereditary Neuropathy with Pressure Palsies (HNPP), respectively (for review, see De Jonghe et al., 1997; Mu¨ller, 2000; Suter and Snipes, 1995). Genotype– phenotype correlations revealed that different mutations cause phenotypes with varying degrees of disease severity (De Jonghe et al., 1997; Nelis et al., 1999; Warner et al., 1996). Furthermore, duplication and deletion of the PMP22 gene is associated with CMT1A and HNPP, respectively, indicating that a gene dosage effect is involved in the pathological mechanism (Patel and Lupski, 1994). Mutations, gene duplication or gene deletion could lead to a nonfunctional protein or interfere with the stoichiometry of protein–protein interactions and ultimately to an unstable myelin structure. PMP22 seems to form heterophilic complexes with P0 in peripheral myelin as well as in ovary carcinoma cells expressing both proteins (D’Urso et al., 1999). Furthermore, PMP22 protein carrying CMT1A-related point mutations accumulates in the ER/Golgi compartments thereby causing a dominant negative effect on the intracellular trafficking of the wild-type PMP22 from the unaffected allele (D’Urso et al., 1998; Naef and Suter, 1999; Naef et al., 1997; Notterpek et al., 1997; Tobler et al., 1999). It has been speculated that the heterophilic interaction of PMP22 and P0 may be important for a function of these proteins in maintenance and stability of PNS myelin (D’Urso et al., 1999; Scherer and Arroyo, 2002). To evaluate the mechanism of the heterophilic PMP22 and P0 interaction, we have generated a cell adhesion assay with retrovirally transduced HeLa cells expressing PMP22 or P0, respectively, and confirmed the results in GST pull down studies. These data indicate that PMP22 and P0 perform homophilic and heterophilic interactions. Mutations located in distinct regions of the extracellular domain of P0 protein that are
B. Hasse et al. / Mol. Cell. Neurosci. 27 (2004) 370–378
associated with CMT1 and DSS disease clearly impair those interactions.
Results Generation of HeLa cell pools expressing P0, PMP22 and Plasmolipin To elucidate the interaction of PMP22 and P0 proteins, we transduced HeLa cells with retroviral constructs encoding PMP22, P0 and Plasmolipin (PLA), respectively. Expression of the constructs in G418 resistant cells was confirmed by reverse transcription (RT)-PCR using construct-specific primers (data not shown) and by Western blotting (Fig. 1). To ensure that each protein was ectopically expressed and inserted into the plasma membrane, intact cells were prelabelled by membrane impermeable biotinylation reagent followed by lysis and subsequent precipitation with Neutravidin. The precipitated material was fractionated by SDS-PAGE, transferred to nitrocellulose membranes, and probed with the specific PMP22, P0 or Plasmolipin antibodies, respectively. Each of the three proteins was expressed at the cell surface of the stably infected HeLa cell pools (Figs. 1A– C) in similar amounts. Adhesive trans-homophilic and trans-heterophilic interactions of PMP22 and P0 To test the ability of the infected cell pools to form homophilic or heterophilic protein interactions, we measured the adhesion of dissociated HeLa cells expressing either PMP22, P0 or Plasmolipin or none of these proteins (nontransduced cells), respectively, to monolayers of cells expressing one of these proteins. Nonadherent cells were removed by rinsing, and the attached cells were stained with MTS for 3 h and analysed photometrically. The adhesion of P0 expressing cells to a monolayer of P0 cells showed an increase in adhesion of 57% ( P b 0.0001; Fig. 2A) when compared with the adherence to a layer of wild-type HeLa cells. Interestingly, cells that overexpressed PMP22 exhibited a 32% increase ( P b 0.001; Fig. 2A) in binding to a monolayer of P0 cells. In contrast, no significant effect was observed when Plasmolipin or noninfected HeLa cells (Fig. 2A), respectively, were measured. Conversely, we
Fig. 1. Western blots of biotinylated plasma membrane proteins from infected HeLa cells. Cell membrane proteins of the infected HeLa cells that overexpressed either PMP22, P0 or Plasmolipin, respectively, were labelled with biotin, lysed and then precipitated with Neutravidin agarose. Precipitates were fractionated by SDS-PAGE, transferred to nitrocellulose membranes, and probed with either anti-PMP22, -P0 or -Plasmolipin antibodies. Detection of P0 (A), PMP22 (B) and Plasmolipin (C) in protein precipitates prelabelled with biotin (+). No signal was obtained in precipitates of infected cells when biotinylation was omitted ( ).
371
detected a similar enhanced adhesion capacity of 43% ( P b 0.001) in the inverted experiment, when P0 cells were added to a monolayer of PMP22 expressing cells (Fig. 2B). Surprisingly, PMP22 overexpressing cells showed a significantly increased homophilic binding capacity of 35% to the PMP22 monolayer ( P b 0.001; Fig. 2B) as compared to the adhesion of Plasmolipin expressing cells or noninfected wild-type HeLa cells (Fig. 2B). However, PMP22 and P0 expressing HeLa cells did not adhere to a monolayer of Plasmolipin expressing cells (Fig. 2C) above noninfected control cells (Fig. 2D), respectively. These results demonstrate that this newly established cell adhesion assay provides a suitable tool to detect homophilic and heterophilic adhesive trans-interactions of membrane proteins. We clearly show that both PMP22 and P0, in contrast to Plasmolipin, perform trans-homophilic as well as trans-heterophilic interactions. The two extracellular domains of PMP22 interact differently in homophilic and heterophilic adhesion To identify protein domains participating in these transinteractions, we purified GST fusion oligopeptides of the extracellular domains of P0, PMP22 and Plasmolipin, respectively, and investigated the adhesive capacity of these fusion proteins in the cell adhesion assay. Wells precoated with specific anti-GST antibodies were coated with P0ex, PMPex1, PMPex2 or PLAex and, as a control, with recombinant GST (rGST). Wells were then incubated either with P0, PMP22 and Plasmolipin (PLA) expressing cells or nontransduced HeLa cells (Wt). As to be expected strong homophilic adhesive interaction of P0 expressing cells to GST-P0 was noted (P0ex, 45%; P b 0.001; Fig. 3A), we also observed a strong adhesion of P0 cells to the second extracellular domain of PMP22 (PMPex2, 65%; P b 0.0001, Fig. 3C) while binding of P0 cells to the first extracellular domain of PMP22 was much weaker (PMPex1, 24%; P b 0.005, Fig. 3B). No binding of P0 cells to PLAex (Fig. 3D) or rGST (Fig. 3E) above control could be observed. Interestingly, PMP22 cells show a very strong and highly significant homophilic binding to the first extracellular domain of PMP22 (PMPex1, 88%; P b 0.0001, Fig. 3B), whereas these cells adhere to the second loop of PMP22 with significantly less affinity (PMPex2, 39%; P b 0.005, Fig. 3C). Furthermore, PMP22 expressing HeLa cells show a strong and highly significant binding to the extracellular domain of P0 (P0ex) (59%; P b 0.0001, Fig. 3A) as compared to the very low adhesion of Plasmolipin cells (15%; P N 0.005, Fig. 3A) or noninfected HeLa cells (Fig. 3A). We never found significant binding of PMP22 or P0 cells to the extracellular domain of Plasmolipin (PLAex; Fig. 3D) or rGST (Fig. 3E) nor significant binding of Plasmolipin cells or nontransduced HeLa cells to either of the investigated protein domains (Figs. 3A–E). Our data prompted us to speculate that the heterophilic trans-interaction of P0 and PMP22 proteins is predominantly mediated through the second extracellular loop of PMP22, whereas the homophilic trans-interaction of PMP22 proteins is largely dependent on the first loop. Direct binding of the extracellular domains of PMP22 to plasma membrane-bound P0 and PMP22 To prove that the cell–protein interaction observed in our adhesion assays were mediated through direct protein–protein interaction between the extracellular domains of PMP22 and
372
B. Hasse et al. / Mol. Cell. Neurosci. 27 (2004) 370–378
Fig. 2. Homophilic and heterophilic adhesion of infected HeLa cells. HeLa cells stably transduced with either PMP22 (HeLaPMP22), P0 (HeLaP0) and Plasmolipin (HeLaPLA), or nontransduced wild-type HeLa cells (HeLaWT) were assayed for adhesion to a monolayer of cells expressing either P0 (A), PMP22 (B), Plasmolipin (C) or to noninfected wild-type HeLa cells (D). Adherent cells were determined by means of a photometric viability assay at 3 h after incubation and removal of nonadherent cells. Proportions of adherent cells are given in [%]. Adherence of infected cell line to wild-type HeLa cell (Wt) monolayer is set to 100% in (A–C). Adherence of HeLaWT to wild-type HeLa cell (Wt) monolayer is set to 100% in D. In each experiment, cell types were assayed in triplicate. Error bars represent the mean F SD of four different experiments (***P b 0.0001, **P b 0.001, and *P b 0.005; Student’s unpaired t test).
P0 with the respective cellular proteins, we carried out a GST pull down assay. Freshly prepared bacterial cell lysates of the GST-tagged extracellular domains of P0ex, PMPex1, PMPex2 and PLAex were bound to glutathione sepharose and the protein-loaded beads were separately incubated with freshly prepared lysates of P0 or PMP22 cells. Bound cellular proteins were fractionated by SDS-PAGE and analyzed by Western blotting. As shown in Fig. 4A, Western blot analyses with an anti-GST antibody confirmed that each fusion protein was coupled to glutathione sepharose beads (GST-P0ex, lanes 1–2; GSTPMPex1, lanes 3–4; GST-PMPex2, lanes 5–6 and GST-PLAex, lanes 7–8). Subsequent Western blot analyses using a P0-specific antibody (Fig. 4B) revealed a direct binding of P0 to its counterpart P0 extracellular domain (GST-P0ex) in homophilic interaction as well as to the second extracellular loop of PMP22 (GST-PMPex2) in heterophilic interaction. No interactions of P0 with either GST-PMPex1 (lane 2) or with GST-PLAex (lane 4) could be detected in this assay (Fig. 4B). On the other hand, Western blot analyses with an anti-PMP22 antibody (Fig. 4C) exhibited a direct association of PMP22 with the extracellular domain of P0 (GST-P0ex, lane 1) in heterophilic binding and with its first loop (GST-PMPex1, lane 2) in homophilic binding.
GST-PMPex2 (lane 3) or GST-PLAex (lane 4) coupled to glutathione sepharose beads did not show PMP22 binding. Mutations within the extracellular domain of P0 affect homophilic and heterophilic interactions Finally, we investigated whether clinically occurring mutations in P0 as described in CMT1B or DSS would affect both the heterophilic and the homophilic adhesion of P0 in our assay system. To this we purified three GST fusion oligopeptides containing the following known mutations of the extracellular domain of P0: Val32Phe (P0mut1), Tyr68Cys (P0mut2) and Cys128Tyr (P0mut3) (Haites et al., 1998; Nelis et al., 1999; Sorour and Upadhyaya, 1998). GST fusion proteins mutated at Val32Phe (P0mut1) and Cys128Tyr (P0mut3) showed a markedly reduced ability to form homophilic interactions with P0 cells (Fig. 5A). The Tyr68Cys mutation (P0mut2), however, only moderately reduced the ability of the extracellular domain of P0 to form homophilic interactions (Fig. 5A). Heterophilic interactions of the P0 domain with PMP22 were nearly abolished by the Cys128Tyr (P0mut3) mutation (Fig. 5B), markedly diminished by the substitution Tyr68Cys (P0mut2) and not affected by the Val32Phe (P0mut1) mutation (Fig. 5B). The data clearly
B. Hasse et al. / Mol. Cell. Neurosci. 27 (2004) 370–378
373
Fig. 3. Adhesion of infected HeLa cells on GST-tagged extracellular protein domains. HeLa cells stably infected with PMP22 (PMP), P0 (P0), Plasmolipin (PLA) and noninfected wild-type HeLa cells (Wt) were assayed for adhesion on purified GST-tagged extracellular domains of (A) P0ex, extracellular domain of P0; (B) PMPex1, first extracellular domain of PMP22; (C) PMPex2, second extracellular domain of PMP22; (D) PLAex, second extracellular domain of Plasmolipin and as control (E) rGST, recombinant Glutathione-S-Transferase. Wells precoated with GST antibodies were coated with 500 ng of each protein and cells were incubated for 2 h. Adherent cells were determined by using a photometric viability assay. Proportions of adherent cells are given in [%]. Adherence of wild-type HeLa (Wt) cells to GST-tagged protein domains is set 100% (A–D), and adhesion of wild-type HeLa (Wt) to rGST is set 100% in (E). In each experiment, cell types were assayed in triplicate. Error bars represent the mean F SD of four different experiments (***P b 0.0001, **P b 0.001, and *P b 0.005; Student’s unpaired t test).
demonstrate significant effects of these P0 mutations on homophilic (P0–P0) and/or heterophilic (P0–PMP22) adhesive trans-interactions.
Discussion The molecular mechanisms of membrane assembly and compaction of the multilamellar structure of myelin sheaths is still
not well understood. Previous studies have identified P0 as a transmembrane protein with homophilic adhesion capacity (D’Urso et al., 1990, 1999; Filbin et al., 1990; Schneider-Schaulies et al., 1990) and, based on crystallography studies it has been suggested that P0 forms interdigitating homotetramers in cis and trans configuration in myelin membranes (Inouye et al., 1999; Shapiro et al., 1996). It was further suggested that the highly basic intracellular domain of P0 could mediate the adhesion of intracellular membrane leaflets in compact myelin through electrostatic
374
B. Hasse et al. / Mol. Cell. Neurosci. 27 (2004) 370–378
Fig. 4. Protein precipitation from infected HeLa cell lysates with GST-tagged extracellular protein domains and Western blotting. The GST tagged extracellular domains of P0, PMP22 and Plasmolipin, respectively, were incubated with HeLa cell lysates overexpressing P0 or PMP22 and subjected to SDS-PAGE followed by Western blot analysis (WB). (A) Detection of the expressed GST tagged extracellular domains of P0 (GST-P0ex, lanes 1–2), PMP22 (GSTPMPex1, lanes 3–4; GST-PMPex2, lanes 5–6) and Plasmolipin (GST-PLAex, lanes 7–8) after the pull down using a monoclonal anti GST antibody. (B) P0 precipitates specifically with P0ex and PMPex2 (lanes 1 and 3). No signals were obtained in fractions containing PMPex1 or PLAex (lanes 2 and 4). (C) PMP22 specifically binds to P0ex and PMPex1 (lanes 1 and 2). Neither PMPex2 nor PLAex (lanes 3 and 4) could precipitate PMP22.
interactions with cytoskeleton proteins and with phospholipids in the membrane bilayer (Ding and Brunden, 1994; Wong and Filbin, 1996). In this investigation, we were able to experimentally confirm the adhesive homophilic trans-interaction of P0 protein through its extracellular domain (Figs. 2A, 3A and 4B). Moreover, we provide evidence by means of three different (cell–cell, cell–protein, protein–protein based) assay systems that P0 also interacts specifically with the second extracellular loop of PMP22. Thus, heterophilic trans-interaction of P0 and PMP22 is likely to contribute to the formation and compaction of the multilamellar myelin sheath in peripheral nerves. Experimental evidence has suggested the formation of intracellular PMP22 homodimers and oligomers in Schwann cells (Ryan et al., 2000; Tobler et al., 1999). However, protein domains mediating homophilic PMP22 adhesion have not been identified previously. Our assay system revealed that PMP22 proteins are also forming adhesive homophilic interactions (Figs. 2B and 3B, C) which, in contrast to the heterophilic P0– PMP22 interaction are mediated through the first extracellular loop of PMP22 (Fig. 4B). Interestingly, no protein–protein interactions could be detected between PMP22 and Plasmolipin or between P0 and Plasmolipin, respectively, indicating that the molecular interactions between PMP22 and P0 proteins are specific. Our data suggest that the homophilic and/or heterophilic transinteractions of integral membrane proteins P0 and PMP22 as characterized in this investigation may play a crucial role in the adhesion of adjacent membrane layers in compact myelin and/or in stabilization of the myelin sheath. Based on the present findings we have modified a recent model (Scherer and Arroyo, 2002) on
protein–protein interactions in peripheral myelin (Fig. 6). It is conceivable that P0 monomers or tetramers (Inouye et al., 1999; Shapiro et al., 1996) may interact in homophilic binding with PMP22 dimers or even oligomers in trans-configuration. Although both P0 and PMP22 carry a L2/HNK1 epitope, which is known to participate in adhesive interactions, glycosylation neither seems to be essential for the homophilic P0–P0 (Griffith et al., 1992; Shapiro et al., 1996) nor for the heterophilic P0–PMP22 interactions (D’Urso et al., 1999). In line with our results is the observation of Tobler et al. (1999) that the N-linked carbohydrate domain is not required for the formation of PMP22 homodimers. On the other hand, for the homophilic adhesion of P0 proteins it was proposed that the sugar moiety is functionally important (Filbin and Tennekoon, 1991, 1993). However, the binding experiments presented in this study were performed with nonglycosylated recombinant proteins derived from bacterial expression systems and with recombinant proteins expressed by cells of non-neural origin. As these cells are unlikely to express P0 or PMP22 proteins with their natural carbohydrate chains, our data suggest that the L2/HNK1 epitope is not required for the initial formation of the homophilic and heterophilic trans-interactions described here. It has been suggested previously that the carbohydrate moiety may play a functional role in stabilizing such a complex (Tobler et al., 1999). Recently, it could be demonstrated that phosphorylation of the cytoplasmic domain of P0 is functionally important for P0mediated cell–cell adhesion (Xu et al., 2001). In our study, we have expressed the P0 extracellular domain (P0ex) as a soluble recombinant protein which lacks both the cytoplasmic and trans-
B. Hasse et al. / Mol. Cell. Neurosci. 27 (2004) 370–378
375
Fig. 5. Effect of mutations in the extracellular domain of P0 on homophilic and heterophilic interactions. (A) Schematic representation of P0. The positions of the three point mutations examined in this study are marked with the respective amino acid substitutions. Figure was modified from Filbin and Zhang (1993). Retrovirally transduced HeLa cells stably exzpressing P0 (B) or PMP22 (C), respectively, were assayed for adhesion on mutated and nonmutated GST-tagged extracellular domains of P0. Mutations used in this assay are Val32Phe (P0mut1), Tyr68Cys (P0mut2) and Cys128Tyr (P0mut3). Ninety-sixwell microplates precoated with GST antibodies were coated with 500 ng of either mutated or nonmutated proteins and cells were incubated for 2 h. Adherent cells were determined by means of a photometric viability assay at 3 h after incubation and removal of nonadherent cells. Adhesion of P0 cells (B) and PMP22 cells (C) to the nonmutated extracellular domain of P0 (P0ex), respectively, was defined as 100%. In each experiment every cell type was assayed in double triplicate. Error bars represent the mean F SD of three different experiments (***P b 0.0001, **P b 0.001, and *P b 0.005; Student’s unpaired t test).
membrane domains. Very interesting, our results show that P0excoated surfaces are sufficient as substrate for the adhesion of P0 cells and for the homophilic P0 interaction. It has been demonstrated previously that the soluble recombinant extracellular domain of P0 could mediate adhesion of P0-expressing cells (Griffith et al., 1992) and homophilic protein interactions in cis and trans (Shapiro et al., 1996). Based on our data, we propose a new model for myelin membrane adhesion that comprises three specific types of protein– protein interactions performed by the integral membrane proteins P0 and PMP22 that include homophilic (P0–P0 and PMP22– PMP22) and heterophilic (P0–PMP22) trans-interactions mediated through distinct extracellular protein domains (Fig. 6). By introducing specific CMT1B- and DSS-related mutations into GST-P0 fusion peptides, we challenged our model and could demonstrate that both the heterophilic and homophilic interactions
Fig. 6. Schematic representation of the three proposed types of homophilic and heterophilic adhesive trans-interactions of integral membrane proteins PMP22 and P0 in peripheral myelin.
376
B. Hasse et al. / Mol. Cell. Neurosci. 27 (2004) 370–378
were differently affected. Amino acid substitution in the N-terminus of P0 (Val32Phe, P0mut1) caused a strong reduction of homophilic binding but had only minor effects on the heterophilic interaction with PMP22, indicating that the N-terminus of P0 is of particular functional importance for P0–P0 trans-interaction (Figs. 5A and B). In contrast, destroying the disulfide bound of P0 (Cys128Tyr and P0mut3) prevented both the homophilic and heterophilc binding, implicating that the formation of the immunoglobulin-like domain is a fundamental functional prerequisite for P0 to form both kinds of interactions. The importance of the immunoglobulin domain of P0 in adhesion is in line with the previous finding that P0 mutated at Cys21 does not behave as a homophilic adhesion molecule (Zhang and Filbin, 1994). Furthermore, we could show that conformational changes in this region caused by introducing an additional cysteine (Tyr68Cys, P0mut2) leads to a stronger reduction in the heterophilic adhesion with PMP22 (Fig. 5C) than in homophilic binding (Fig. 5B), indicating that the heterophilic interaction is more sensitive to alterations or modifications of the immunoglobulin domain than the homophilic interaction. Our findings provide strong evidence that alterations in P0 protein which interfere with the formation of homophilic and heterophilic protein complexes could destabilize the structure of myelin, thus resulting in pathological phenotypes such as CMT1B or the more severe disease form DSS. The results further demonstrate that the adhesion assay applied in this investigation is sensitive enough to detect specific differences in the protein binding capacity of the wild-type and mutated extracellular domains of P0. While the most affected alteration in protein binding was associated with the Cys128Tyr substitution in P0, a mutation found in a severe DSS disease phenotype, the two other mutations tested in this study (Val32Phe and Tyr68Cys) were associated with milder CMT1B phenotypes corresponding to minor effects of these genotypes on alterations in P0 binding properties. Our present observations provide new insight into the specific molecular interactions of the two major peripheral myelin proteins PMP22 and P0. The data suggest that domain-specific homophilic and heterophilic adhesive trans-interactions of these proteins are likely to play a role in the mechanism that determines the molecular architecture and stability of compact myelin sheaths in peripheral nerve. In addition, we have demonstrated that CMT/ DSS-related point mutations of P0 resulted in decreased adhesion capabilities corresponding with the severity of the respective disease phenotype.
Experimental methods Antibodies The following antibodies were used as indicated: Polyclonal anti-PMP22 antibody (1:300; Antibody Service, Dr. Pineda, Berlin, Germany), polyclonal anti-P0 antibody (1:500; Hasse et al., 2002), polyclonal anti-Plasmolipin antibody (1:200; Hamacher et al., 2001) and monoclonal anti-Glutathione-S-Transferase antibody (1:2000; Amersham Pharmacia Biotechnology, Buckinghamshire, UK).
EGFP cassette in the 5V multicloning site of the retroviral plasmid S11IEG3 (data not shown). To allow selection of transduced cells the EGFP-coding region of the IRES-EGFP cassette was replaced by an ORF with the neomycin phosphotransferase gene. The PCR amplification products and all relevant cloning sites of the resulting retroviral S11PMP22IN, S11P0IN and S11PlasmolipinIN plasmids were sequenced using an ABI310 sequencer (Applied Biosystems, Weiterstadt, Germany). For generation of stable packaging cell lines ecotropic enveloped-pseudotyped retroviral supernatants were generated in each case by transient transfection of subconfluent ecotropic Phoenix packaging (Nolan Lab, Stanford; Schulze et al., 2002) with 15 Ag of plasmid DNA and 45 Ag of Fugene (Roche, Mannheim, Germany) in 100-mm culture dishes. The next day DMEM culture medium was replaced by IMEM medium and 2 days after transfection retrovirus containing medium collected and filtered through 0.45 Am filters (Sartorius, Gfttingen, Germany). These ecotropically pseudotyped retroviruses were used to infect PG13-packaging cells (Miller et al., 1991) for 3 days in the presence of 7.5 Ag/ml of protaminsulfate (Sigma, Deisenhofen, Germany)/ml medium (Hanenberg et al., 2002). Culture supernatants from transduced PG13 cells containing GALV-pseudotyped retroviral vectors were harvested from confluent dishes every 24 h for 4 days, purified by filtration through 0.45 Am and stored at 808C until use. Retroviral transductions of subconfluent HeLa cells (40–60%) with the infectious PG13-supernatants were performed in the presence of 7.5 Ag protaminsulfate two times for 24 h. To generate stably overexpressing HeLa cell populations, the HeLa cells were replated the next day and cultured in the presence of 0.8 mg/ml Geniticin (Sigma) for at least 10 days. PMP22, P0 and Plasmolipin overexpression of the generated cell pools was confirmed by RTPCR (data not shown) and Western blot analysis, respectively. Cell surface biotinylation and Western blot analyses To analyze protein expression of infected HeLa cell pools and to determine plasma membrane localisation of the overexpressed proteins, cell surface molecules were labelled by biotinylation of intact cells. Confluent cell layers (4 106 cells/pool) in 100-mm dishes were rinsed free of serum and labeled with 0.5 mg/ml freshly prepared membrane impermeable Sulfo-NHS-LC-Biotin or PEO-Maleimiide Activated Biotin (both supplied by Pierce Biotechnology, Rockford, IL, USA) in ice cold PBS for 30 min at room temperature. After washing and quenching, cell lysates were prepared using 1.5 ml of lysis buffer (Pierce Biotechnology) per dish and cleared by centrifugation at 12,000 g for 5 min. Supernatant was precipitated with Neutravidin-coupled agarose (Pierce Biotechnology) for 1 h at 48C. After several washes with lysis buffer, beads were resuspended in 2% SDS sample buffer and boiled for 5 min. Biotinylated cell surface proteins were fractionated by SDS-PAGE, transferred to Nitrocellulose membranes and probed with the specific antibodies indicated above followed by HRP-conjugated antirabbit IgG or antigoat IgG. Signals were visualized using a chemiluminescent detection system (ECL; Amersham Biotechnology).
Retroviral overexpression of PMP22, P0 and Plasmolipin in vitro Cloning of wild-type and mutated GST-tagged fusion oligopeptides The open reading frames (ORFs) of rat PMP22 (NM017037), rat P0 (K03242, BF562392) and rat Plasmolipin (PLA, NM022533) were directionally inserted in front of the IRES-
The rat wild-type cDNAs of PMP22, P0 and Plasmolipin in pIRES2-EGFP (Hasse et al., 2002) were used as templates to gen-
B. Hasse et al. / Mol. Cell. Neurosci. 27 (2004) 370–378
erate the extracellular domain of P0 (P0ex, residues 30–153), the second extracellular domain of Plasmolipin (PLAex, residues 121– 141) and both extracellular domains of PMP22 (PMPex1, residues 32–64 and PMPex2, residues 120–133). The PCR constructs were amplified using the following primers: sense P0ex 5V-CCAGGAATTCGTATTGTGGTTTACACGGAC-3V and antisense 5V-GCATG T C G A C C T A G T G G G C A C - 3 V, s e n s e P L A e x 5 VCCAGGAATTCGTG CCTGTGCGGCGGCGGTC-3V and antisense 5V-CCAGGAATTCGTAATGGACACA GGACTGAT-3V; sense PMPex1 5V-GCATGTCGACCTAAGACTGAAGCCATTCGCT-3V and antisense 5VGCATGCGACCTAAGACTG A A G C C AT T C G C T- 3 V s e n s e P M P e x 2 5 V- C C A G G Q AATTCGTAGACACAGTGAGTGGCATGTC-3V and antisense 5V-GCATGT CGACCTAAAAGCCATAGGAGTAGT-3V. The cDNAs encoding the P0ex mutants were constructed using PCR-mediated mutagenesis. The following three-point mutations were applied in this study: Val32Phe (P0mut1), Tyr68Cys (P0mut2) and Cys128Tyr (P0mut3). All these amino acid substitutions have been reported in human CMT1 (Val32Phe and Tyr68Cys) and Dejerine Sottas Syndrome (Cys128Tyr), respectively, with affected myelin sheaths (Haites et al., 1998; Nelis et al., 1999; Sorour and Upadhyaya, 1998). The constructs were amplified using the nonmutated primers and the following mutated primers (mutations were underlined): sense P0mut1 5V-CCAGGAATCGATTGTGTTTTACACGGAC-3V and antisense P0mut1 5V-GCATGTCGACCTAATACCTAGTGGGCAC-3V; sense P0mut2 5V-TCT TTTACCTGGCGCTGCCAGCCT-3V and antisense P0mut2 5V-AGGCTGGCAGCGC-CAGGT-3V, sense P0mut3 5V-CAACGGCACTTTCACATATGATGTC and antisense P0mut3 5V-GACATCATATGTGAAAGTGCCG-3V. All wild-type and mutated constructs of the extracellular domains were ligated in frame into the EcoR1/Sal1 sites of the Glutathione-S-Transferase (GST) gene fusion vector pGEX-4T2 (Amersham Pharmacia). To prove the identity of the expression constructs, the inserted cDNAs were sequenced using the dideoxy chain termination method on the ABI Prism 310 (Applied Biosystems). Purification of recombinant proteins To express the GST-tagged extracellular domains of P0, PMP22 and Plasmolipin, the plasmids were transformed into Escherichia coli strain BL21 (Amersham Pharmacia Biotechnology). Bacterial cells were induced with 0,5 mM IPTG for protein overexpression and the bacterial proteins were solubilized according to the manufacturer’s protocol. Recombinant GST-tagged proteins were purified using Glutathione affinity chromatography. The purity of the GST-tagged extracellular domains of P0, PMP22 and Plasmolipin was proven by Coomassie Blue staining (data not shown) and western blotting.
377
separately. After incubation for 2 h at 378C, nonadherent cells were removed by rinsing the cultures vigorously several times. The numbers of cells adhering to the monolayers were detected using the CellTiter 96 Aqueous Cell Proliferation Assay (Promega, Bochum, Germany). In brief, cells were incubated with a methyltetrazolium salt (MTS) which was reduced by living cells to a soluble formazan product. The resultant soluble formazan salt was equivalent to the number of living active cells and was quantified by measuring the absorbance at 490 nm with a microtiter plate reader after 3 h. Additionally, to determine the number of adhering cells the overnight monolayers of each cell population were measured separately and subtracted as background. To determine the cell adhesion on purified wild-type or mutated GST-tagged extracellular domains of P0, PMP22 and Plasmolipin, respectively, 96-well micro plates precoated with antiGST antibody (Amersham Pharmacia, GST Detection Kit) were incubated with 500 ng of the purified proteins and, as control, with recombinant GST overnight at 48C. After several washes, HeLa cells of each cell pool were plated onto the proteins at 15 103 cells/well for 2 h at 378C. Nonadherent cells were carefully washed away and remaining cells were assayed for adhesion as described above. Binding to each protein or cell monolayer was assayed in triplicate. Each experiment was repeated at least four times. An unpaired Student’s t test was used to confirm significance of the results. GST pull down assay Freshly prepared bacterial cell lysates of the GST-tagged extracellular domains of P0, PMP22 and Plasmolipin, respectively, were bound to glutathione Sepharose for 1 h at 48C. For precipitation of interacting partners, stably infected P0 or PMP22 HeLa cells were solubilized in lysis buffer and shaken for 30 min at 48C. To avoid nonspecific binding of proteins in the cellular lysates, the lysates were preadsorbed with GST-Sepharose beads for 1 h and then incubated over night at 48C with the GST-tagged extra cellular protein domains linked to Sepharose beads. The beads were pelleted at 1250 g for 2 min and washed five times with lysis buffer. Protein complexes of GST-tagged extracellular domains and proteins from HeLa cell lysates were resuspended in SDS sample buffer and boiled for 5 min. Samples were analyzed by Western blotting as described above.
Acknowledgments This work was supported in part by the Deutsche Forschungsgemeinschaft (Mu630/5-4), the Elterninitiative Kinderkrebsklinik and a Grant from the Medical Faculty of the University of Dqsseldorf.
Cell adhesion assay Appendix A. Supplementary data Cell layers of wild-type HeLa cells or either PMP22, P0 and plasmolipin expressing cells, respectively, were washed free of serum, harvested using 0,1% trypsin in PBS buffer and resuspended in DMEM supplemented with 10% fetal calf serum, 100 U/ ml penicillin and streptomycin, and 2 mM l-glutamine (all supplied by Gibco BRL, Grand Island, NY) and plated on 96well microplates at 5 103 cells/well overnight. To each of these cell layers additional 5 103 cells of each cell pool, resuspended in DMEM supplemented with 20 mM EDTA, were added
Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.mcn.2004.06.009.
References Arroyo, E.J., Scherer, S.S., 2000. On the molecular architecture of myelinated fibers. Histochem. Cell Biol. 113, 1 – 18.
378
B. Hasse et al. / Mol. Cell. Neurosci. 27 (2004) 370–378
De Jonghe, P., Timmerman, V., Nelis, E., Martin, J.J., van Broeckhoven, C., 1997. Charcot–Marie–Tooth disease and related peripheral neuropathies. J. Peripher. Nerv. Syst. 2, 370 – 387. Ding, Y., Brunden, K.R., 1994. The cytoplasmic domain of myelin glycoprotein P0 interacts with negatively charged phospholipid bilayers. J. Biol. Chem. 269, 10764 – 10770. D’Urso, D., Brophy, P.J., Staugaitis, S.M., Gillespie, C.S., Frey, A.B., Stempak, J.G., Colman, D.R., 1990. Protein zero of peripheral nerve myelin: biosynthesis, membrane insertion, and evidence for homotypic interaction. Neuron 4, 449 – 460. D’Urso, D., Prior, R., Greiner-Petter, R., Gabreels-Festen, A.A., Mqller, H.W., 1998. Overloaded endoplasmic reticulum–Golgi compartments, a possible pathomechanism of peripheral neuropathies caused by mutations of the peripheral myelin protein PMP22. J. Neurosci. 18, 731 – 740. D’Urso, D., Ehrhardt, P., Mqller, H.W., 1999. Peripheral myelin protein 22 and protein zero: a novel association in peripheral nervous system myelin. J. Neurosci. 19, 3396 – 3403. Filbin, M.T., Tennekoon, G.I., 1991. The role of complex carbohydrates in adhesion of the myelin protein Po. Neuron 7, 845 – 855. Filbin, M.T., Tennekoon, G.I., 1993. Homophilic adhesion of the myelin Po protein requires glycosylation of both molecules in the homophilic pair. J. Cell Biol. 122, 451 – 459. Filbin, M.T., Zhang, K., 1993. The Ig-disulfide bond of myelin Po protein is essential to its adhesion. J. Neurochem. 61, 194. Filbin, M.T., Walsh, F.S., Trapp, B.D., Pizzey, J.A., Tennekoon, G.I., 1990. Role of myelin P0 protein as a homophilic adhesion molecule. Nature 344, 871 – 872. Greenfield, S., Brostoff, S., Eylar, E.H., Morell, P., 1973. Protein composition of myelin of the peripheral nervous system. J. Neurochem. 20, 1207 – 1216. Griffith, L.S., Schmitz, B., Schachner, M., 1992. L2/HNK-1 carbohydrate and protein–protein interactions mediate the homophilic binding of the neural adhesion molecule P0. J. Neurosci. Res. 33, 639 – 648. Haites, N.E., Nelis, E., Van Broeckhoven, C., 1998. 3rd Workshop of the European CMT consortium: 54th ENMC International Workshop on Genotype/Phenotype Correlations in Charcot–Marie–Tooth Type 1 and Hereditary Neuropathy with Liability to Pressure Palsies 28–30 November 1997, Naarden, The Netherlands. Neuromuscular Disord. 8, 591 – 603. Hamacher, M., Pippirs, U., Kohler, A., Mqller, H.W., Bosse, F., 2001. Plasmolipin: genomic structure, chromosomal localization, protein expression pattern, and putative association with Bardet–Biedl syndrome. Mamm. Genome 12, 933 – 937. Hanenberg, H., Batish, S.D., Pollok, K.E., Vieten, L., Verlander, P.C., Leurs, C., Cooper, R.J., Gottsche, K., Haneline, L., Clapp, D.W., Lobitz, S., Williams, D.A., Auerbach, A.D., 2002. Phenotypic correction of primary Fanconi anemia T cells with retroviral vectors as a diagnostic tool. Exp. Hematol. 30, 410 – 420. Hasse, B., Bosse, F., Mqller, H.W., 2002. Proteins of peripheral myelin are associated with glycosphingolipid/cholesterol-enriched membranes. J. Neurosci. Res. 69, 227 – 232. Inouye, H., Tsuruta, H., Sedzik, J., Uyemura, K., Kirschner, D.A., 1999. Tetrameric assembly of full-sequence protein zero myelin glycoprotein by synchrotron x-ray scattering. Biophys. J. 76, 423 – 437. Miller, A.D., Garcia, J.V., von Suhr, N., Lynch, C.M., Wilson, C., Eiden, M.V., 1991. Construction and properties of retrovirus packaging cells based on gibbon ape leukemia virus. J. Virol. 65, 2220 – 2224. Mqller, H.W., 2000. Tetraspan myelin protein PMP22 and demyelina-
ting peripheral neuropathies: new facts and hypotheses. Glia 29, 182 – 185. Naef, R., Suter, U., 1999. Impaired intracellular trafficking is a common disease mechanism of PMP22 point mutations in peripheral neuropathies. Neurobiol. Dis. 6, 1 – 14. Naef, R., Adlkofer, K., Lescher, B., Suter, U., 1997. Aberrant protein trafficking in Trembler suggests a disease mechanism for hereditary human peripheral neuropathies. Mol. Cell. Neurosci. 9, 13 – 25. Nelis, E., Haites, N., van Broeckhoven, C., 1999. Mutations in the peripheral myelin genes and associated genes in inherited peripheral neuropathies. Hum. Mutat. 13, 11 – 28. Notterpek, L., Shooter, E.M., Snipes, G.J., 1997. Upregulation of the endosomal–lysosomal pathway in the trembler—J neuropathy. J. Neurosci. 17, 4190 – 4200. Pareek, S., Suter, U., Snipes, G.J., Welcher, A.A., Shooter, E.M., Murphy, R.A., 1993. Detection and processing of peripheral myelin protein PMP22 in cultured Schwann cells. J. Biol. Chem. 268, 10372 – 10379. Patel, P.I., Lupski, J.R., 1994. Charcot–Marie–Tooth disease: a new paradigm for the mechanism of inherited disease. Trends Genet. 10, 128 – 133. Ryan, M.C., Notterpek, L., Tobler, A.R., Liu, N., Shooter, E.M., 2000. Role of the peripheral myelin protein 22 N-linked glycan in oligomer stability. J. Neurochem. 75, 1465 – 1474. Scherer, S.S., Arroyo, E.J., 2002. Recent progress on the molecular organisation of myelinated axons. J. Peripher. Nerv. Syst. 7, 1 – 12. Schneider-Schaulies, J., von Brunn, A., Schachner, M., 1990. Recombinant peripheral myelin protein P0 confers both adhesion and neurite outgrowth-promoting properties. J. Neurosci. Res. 27, 286 – 297. Schulze, K.M., Hanemann, C.O., Mqller, H.W., Hanenberg, H., 2002. Transduction of wild-type merlin into human schwannoma cells decreases schwannoma cell growth and induces apoptosis. Hum. Mol. Genet. 11, 69 – 76. Shapiro, L., Doyle, J.P., Hensley, P., Colman, D.R., Hendrickson, W.A., 1996. Crystal structure of the extracellular domain from P0, the major structural protein of peripheral nerve myelin. Neuron 17, 435 – 449. Sorour, E., Upadhyaya, M., 1998. Mutation analysis in Charcot–Marie– Tooth disease type 1 (CMT1). Hum. Mutat. (Suppl. 1), S242 – S247. Suter, U., Snipes, G.J., 1995. Peripheral myelin protein 22: facts and hypotheses. J. Neurosci. Res. 40, 145 – 151. Tobler, A.R., Notterpek, L., Naef, R., Taylor, V., Suter, U., Shooter, E.M., 1999. Transport of Trembler-J mutant peripheral myelin protein 22 is blocked in the intermediate compartment and affects the transport of the wild- type protein by direct interaction. J. Neurosci. 19, 2027 – 2036. Warner, L.E., Roa, B.B., Lupski, J.R., 1996. Absence of PMP22 coding region mutations in CMT1A duplication patients: further evidence supporting gene dosage as a mechanism for Charcot–Marie–Tooth disease type 1A. Hum. Mutat. 8, 362 – 365. Wong, M.H., Filbin, M.T., 1996. Dominant-negative effect on adhesion by myelin Po protein truncated in its cytoplasmic domain. J. Cell Biol. 134, 1531 – 1541. Xu, W., Shy, M., Kamholz, J., Elferink, L., Xu, G., Lilien, J., Balsamo, J., 2001. Mutations in the cytoplasmic domain of P0 reveal a role for PKCmediated phosphorylation in adhesion and myelination. J. Cell Biol. 155, 439 – 446. Zhang, K., Filbin, M.T., 1994. Formation of a disulfide bond in the immunoglobulin domain of the myelin P0 protein is essential for its adhesion. J. Neurochem. 63, 367 – 370.
