Species-Specific Recognition Patterns of Monoclonal Antibodies Directed against Vimentin BoHN,**~ WOLFRAM WIEGERS,? MICHAEL BEUTTENM~~LLER,~ AND PETER TRAUB~
Experimentelle Virologie und Zmmunologie an der Universittit Hamburg, D-2000 Hamburg and tMar-Planck-Znstitut fiir Zellbiologie, D-6802 Ladenburg, Germany
[2, 31. The sequence homologies are most pronounced within the central rod domain, the site by which these proteins associate laterally into tetrameric protofilament units [4-61. Sequence homologies within this rod domain provided the basis for dividing the IF proteins into five sequence types [ 71. Vimentin, a type III IF protein of 53 kDa, represents the characteristic IF subunit in cells of mesenchymal origin . Comparison of chicken and mouse vimentin amino acid sequences revealed an overall identity of about 80%, with close to 100% identity within the rod domain . Corresponding to this strongly conserved structure vimentin antibodies usually react with cells of a wide range of animal species . This was a great advantage in checking for the presence of vimentin in the tissue of various animal species and favored the use of these antibodies for classification of tumors in man and experimental animal studies [lo121. However, the non-species-specific reaction considerably impairs immunocytochemical investigations of expression of heterologous vimentin proteins in cultured cells [13, 141. Vimentin antibodies that show a more restricted reaction pattern can overcome these difficulties in experimental studies [ 151. Here we report that two commercially available antibodies are able to 0 1992 Academic Press, Inc. distinguish between vimentin from different species. The epitope recognized by one of these antibodies (VIM 3B4) has been characterized more precisely and is INTRODUCTION shown to encompass a region around residue 353 of the rod domain of murine vimentin. The specific binding Intermediate filaments (IFS) comprise a family of pattern of this antibody strongly suggests that rodent cytoskeletal proteins in animal cells, whose expression vimentin contains a common residue at this position is tissue-specifically and developmentally regulated [ 11. which specifically differs from that present in vimentin Biochemical characterization and sequence comparison of other mammalian orders. have revealed a three-domain structure common to all IF proteins, namely an a-helix rich central rod domain MATERIAL AND METHODS of approximately 310 residues flanked by two nonhelical Cells. The following cell lines were used in the studies: human, domains, the amino-terminal head domain and the carHiKa , HeLa (ATCC CCLZ), and MRC-5 (ATCC CCL 171); Afriboxy-terminal tail domain. Corresponding to this comcan green monkey, Vero (ATCC CCL 81), CVl (ATCC CCL 70), and mon building scheme IF proteins reveal substantial ho- COS-7 (ATCC CRL 1651); rhesus monkey, LLC-MK, (ATCC CCL7); mologies in amino acid sequence and secondary struccanine, MDCK (ATCC CCL34); mouse, 3T3, skin fibroblasts (MSF), F9 (ATCC CRL 1720); rat, C, glial cell (ATCC CCL 107), fibroblasts ture and form filaments of identical shape and diameter
Two commercially available monoclonal antibodies raised against the intermediate filament protein vimentin were characterized concerning their species-specific reaction pattern on vertebrate cells. The antibody VS exhibited extensive reactivity with vimentin of all mammalian species tested, but specifically did not detect vimentin in mouse cells and chicken fibroblasts. The antibody VIM 3B4 recognized vimentin in cells of chicken and most mammalian species, except for rodent species. Characterization of the binding site of VIM 3B4 on human vimentin by limited proteolysis and immunoblotting as well as by sequence comparison strongly suggested that the epitope is located in the coil 2 part of the vimentin rod domain. Site-directed mutagenesis of a mouse vimentin cDNA clone followed by in vivo expression showed that VIM 3B4 could detect rodent vimentin containing a single amino acid substitution (valine for leucine) at position 353 of the mouse vimentin sequence. Practical application for this finding was demonstrated by the unequivocal identification of a modified murine vimentin protein, distinct from the endogenous vimentin, in a cytoplasmic intermediate filament network in mouse skin fibroblasts transfected with a recombinant plasmid expression vector.
’ To whom reprint
(RAF), and NRK-52E (A’I’CC CRT, 1571); Syrian hamster, BHK-21 (gift from Sandoz, Vienna); Chinese hamster, CHO; gerbil, IMR-33 (ATCC CCL 146); rabbit, SD Ep (ATCC CCL 68); bovine, EBTr
requests should be addressed. 1
0014-4827/92 $5.00 Copyright G 1992 by Academic Press, Inc. rights of reproduction in any form reserved.
(ATCC CCL 44); equine, E. Derm (ATCC CCL 57); and chicken, primary fibroblasts. The cells were grown in minimal essential Eagle’s medium supplemented with 10% fetal calf serum. Cells of the subclone F8  of the human adrenal cortex carcinoma-derived cell line SW 13  were grown in L-15 medium supplemented with 10% FCS. Indirect immunofluorescence microscopy. Cells were grown on glass coverslips and fixed with cold acetone (-20°C) for 10 min. Alternatively, the cells were prefixed in phosphate-buffered saline containing 3.7% formaldehyde for 10 min at room temperature and permeabilized for 2 min in phosphate-buffered saline containing 3.7% formaldehyde and 0.2% Triton X-100. Vimentin was localized with either the monoclonal antibodies clones V9  and VIM 3B4  (Boehringer-Mannheim, Mannheim, Germany) at recommended dilutions of 1:4 and 1:5, respectively, or an affinity-purified antibody to vimentin raised in goat . Affinity-purified goat anti-mouse-FITC antibodies (Sigma, Deisenhofen, Germany) and rabbit anti-goat-TRITC antibodies (DAKO, Hamburg, Germany) were used as secondary antibodies. Zmmunoblotting. Cells were grown to confluence in 25-cm2 culture flasks. They were lysed with sample gel buffer containing 2% SDS and 5% mercaptoethanol. The samples were sonicated for 10 s and heated to 100°C for 5 min. Twenty microliters of each lysate was run on a 10% polyacrylamide slab gel. The proteins were transferred onto nitrocellulose filters by semidry blotting. Nonspecific binding sites on the nitrocellulose sheets were blocked with 1% bovine serum albumin in 0.2 M phosphate buffer (pH 7.4). The sheets were incubated with the monoclonal vimentin antibodies V9 and 3B4 (working dilution 1:50) or with the affinity-purified goat antibodies to vimentin (1:lOO) for 3 h at room temperature. After washing with buffer containing 0.1% Tween 20 the sheets were reacted with peroxidase-conjugated rabbit anti-mouse (DAKO) or rabbit anti-goat (Jackson, Dianova, Hamburg) antibodies. The bound antibodies were detected with chloronaphthol according to Hawkes et al. . Proteolytic degradation ofvimentin. Human vimentin was isolated from HeLa cells by single-stranded DNA-cellulose affinity chromatography according to Nelson et al. . The purified vimentin protein was subjected to limited proteolysis using recombinant human immunodeficiency virus type 1 protease (HIV-l PR) . Cleavage products were separated on a 12% polyacrylamide slab gel, fixed, and stained with Coomassie blue. Alternatively, the proteins were blotted onto nitrocellulose and characterized with vimentin antibodies according to Shoeman et al. . A full-length mouse Site-directed mutagenesis of vimentin cDNA. vimentin cDNA clone previously described  was subcloned into phage M13MP19 RF- (replicative form) DNA. Base C in codon CTT, coding for leucine-353 , was changed to base G thereby resulting in codon GTT, coding for valine, by oligonucleotide-directed in vitro mutagenesis using a synthetic oligonucleotide (GAA GAG AAT TTT GCC GTT GAA GCT GCT) and the Muta-Gene Ml3 in vitro mutagenesis kit (Bio-Rad, Munich, Germany [27,28]). Mutants were identified by sequencing the single-stranded M13MP19 DNA using the Sequenase V2.0 sequencing kit (USB, Cleveland). A DNA fragment containing the desired mutation was subcloned into the eukaryotic expression vector pSG5 (Stratagene, La Jolla), which drives expression of the inserted cDNA by the SV-40 early promoter. The resulting plasmid was designated pVIM’“‘353. Microinjection. Microinjection was performed as previously described . Cells were microinjected with either the recombinant plasmid expression vector pVIMVe’353 at a concentration of 0.1 mg/ml in 5 mA4 3-(N-morpholino)propanesulfonic acid (pH 7) or with human vimentin protofilaments (0.5 mg/ml) and processed for indirect immunofluorescence at various times after injection.
Recognition Patterns of Monoclonal Antibodies Initial immunofluorescence labeling with the monoclonal antibodies V9 and VIM 3B4 done on acetone-
fixed human fibroblasts (not shown) and Vero (green monkey) cells (Figs. la and lb) showed the typical staining pattern of vimentin filaments being largely concentrated around the nucleus. Immunofluorescence labeling on C6 rat glioma cells (not shown) and BHK-21 cells (Figs. lc and Id) again was positive with the monoclonal V9 antibody and with a polyclonal vimentin antibody, whereas the monoclonal antibody VIM 3B4 failed to detect any vimentin filaments in these cells. These results were consistently obtained, irrespective of the fixation protocol employed. This suggested that the VIM 3B4 antibody could distinguish between human and rat vimentin, at least in immunofluorescence. Subsequently, cell lines of various species were tested with the V9 and VIM 3B4 antibodies by immunofluorescence as is summarized in Table 1. Obviously, the V9 antibody recognized vimentin in cells of human, rhesus and african green monkey, canine, equine, bovine, rabbit, rat, and hamster origin, but did not react with vimentin in mouse cells and chicken fibroblasts. The VIM 3B4 antibody also detected vimentin in primate cells and cells of canine, rabbit, equine, and bovine, but generally failed to detect vimentin in cells of rodent origin. In addition this antibody recognized vimentin in chicken fibroblasts (Fig. If). In all of these cells, the polyclonal antibody to vimentin showed positive staining at least in immunofluorescence (data not shown). Immunoblotting The immunofluorescence data suggested that the epitope detected by VIM 3B4 on primate vimentin was either not present or not accessible on vimentin of rodent cells. Thus Western immunoblotting was performed to evaluate whether this epitope was differentially masked in these cells. Lysates of CVl (primate), BHK-21 (hamster), NRK (rat), and chicken fibroblasts cells were run on a SDS-polyacrylamide slab gel under reducing conditions, and the blotted proteins were reacted with either the monoclonal or polyclonal anti-vimentin antibodies (Fig. 2). In summary, the V9 antibody detected vimentin in the lysates of primate, rat, and hamster cells but not of chicken cells. In contrast, the VIM 3B4 antibody localized this protein in lysates of primate and chicken cells but not of rodent cells. Sequence Comparison
Further studies were performed to localize the epitope recognized by the VIM 3B4 antibody more precisely. Vimentin was purified from human (HeLa) cells and subjected to limited proteolysis by the HIV-l protease. It is known that cleavage of human vimentin by this protease results in the production of the intact (38 kDa) rod domain, corresponding to amino acids 93 to 422, and a number of smaller peptides [ 241. Immunoblotting of the cleavage products showed that the VIM 3B4
VIM 3 B4
FIG. 1. Immunofluorescence microscopy of Vero cells (a and b), BHK-21 cells (c and d), and chicken fibroblasts the V9 (a, c, and e) or the VIM 3B4 (b, d, and f) monoclonal antibody. Magnification, 500X.
antibody recognized this rod domain fragment of vimentin, as well as larger proteolytic fragments and the intact vimentin protein (Fig. 3). These data suggested that the epitope of VIM 3B4 is localized in the rod domain of human vimentin. Comparison of the cDNA nucleotide and derived amino acid sequences of vimentin from man, mouse, hamster, and chicken (see Ref. ) revealed five differences in the amino acid sequence of the rod domain fragment (aa 93-422) between mouse and human vimentin (Table 2). In each of these exchange positions the residues originally given for mouse  and hamster  were identical, but differed from those in human
(e and f) employing
vimentin . Amino acid 353 in human vimentin was identical to the corresponding residue in chicken vimentin, which would match with the labeling pattern of VIM 3B4. The data strongly argued that a single amino acid substitution at position 353 of mouse vimentin prevented binding of the VIM 3B4 antibody. The correctness of this assumption was checked by site-directed mutagenesis. We changed the “murine” leucine at position 353 to the “human” valine in the full-length murine20vimentin cDNA and cloned this sequence into a constitutive expression vector. Using microinjection, MSF and human SW 13 F8 cells (which are devoid of cytoplasmic IFS) were then transfected with
V9 and VIM3B4 Recognition
Carnivora Perrissodactyla Artiodactyla Lagomorpha Rodentia
Rhesus monkey Canine Equine Bovine Rabbit Gerbil Murine
Hamster Class aves
this plasmid (PVIM’“‘~~~). After 20 h, cells were fixed and labeled with the VIM 3B4 antibody. Transfected mouse cells expressing the mutated mouse vimentin (Fig. 4b) showed the typical architecture of a wellspread interphase network in immunofluorescence, while endogenous vimentin in the surrounding nontransfected cells revealed no labeling. Similar results were obtained with transfected SW13 F8 cells which contain no endogenous vimentin (Figs. 4c and 4d). In MSF cells injected with human vimentin protofilaments the VIM 3B4 antibody could clearly localize these subunits, which were still unassembled, whereas the en-
+ + + f + + + + + + + + + + + + + -
HiKa MRC-5 HeLa Vero CVl cos-7 LLCMKP MDCK E. Derm EBtr. Sfl Ep. IMR-33 3T3 F9 MSF NRK C6 RAF BHK-21 CHO Primary fibroblasts
+ + + + + + + + + + + +
dogenous filamentous vimentin network remained unlabeled (Fig. 4a). DISCUSSION
The immunocytochemical and biochemical data provide evidence for a species-specific vimentin recognition pattern of the two monoclonal antibodies V9 and VIM
FIG. 2. Western immunoblotting. Lysates of cells were run on a 10% SDS-polyacrylamide slab gel, blotted onto nitrocellulose sheets, and reacted with the monoclonal vimentin antibodies V9, VIM 3B4, or with a polyclonal vimentin antibody (pV).
FIG. 3. SDS-PAGE of human vimentin protofilaments digested by HIV-l protease. Digestion was stopped after 0 min (lanes a), 2 min (lanes b), 5 min (lanes c), or 30 min (lanes d). Protein standards (SDS 7B; Sigma) were run in lane M. Proteins were separated on 12% polyacrylamide SDS gels and were either stained with Coomassie blue (A) or electrotransferred to nitrocellulose sheets for labeling with the VIM 3B4 antibody (B).
Comparison of Amino Acid Differences in the Vimentin Rod Domain Fragment of Different Species * Mouse Hamster Human Chicken
201 200 201 195
Ser Ser Asn Ser
239 Asp 238 Asp 239 Glu 233 Asp
317 316 317 311
Asn Asn Thr Asn
353 352 353 347
Leu 417 Thr Leu 416 Asn Val417 Asn Val411 Thr
Note. Amino acid positions refer to sequence comparison data given in Ref. ; for primary sequence data see the following references: mouse , hamster , human [31, 321, and chicken . * Position of mutagenesis site.
3B4. The antibody V9 specifically did not detect vimentin in mouse cell lines, in agreement with previous findings , whereas the antibody VIM 3B4 generally failed to detect this IF protein in cells of rodent species. Both antibodies reacted with vimentin in cells of all other mammalian species tested, but differed in their
binding to chicken vimentin. The reaction pattern on mammalian cells suggests that rodent vimentin contains specific sites which are either unique to murine species or are characteristic of the whole order. Similar to VIM 3B4, a monoclonal antibody that failed to bind to vimentin in 3T3 mouse cells and reacted with chicken fibroblasts was described by others 1151. Whether this antibody recognizes vimentin of other rodent and mammalian species was not determined. The recognition pattern of V9 and VIM 3B4 was also evident after denaturing the cellular proteins with SDS followed by gel electrophoresis and immunoblotting. This makes it unlikely that the epitopes recognized by these antibodies are differentially masked in the various cell lines. The failure of the antibodies to localize vimentin in certain cells rather documents the absence of a corresponding epitope, a suggestion which could be substantiated with respect to the VIM 3B4 epitope. Limited proteolysis of human vimentin by HIV-l PR in conjunction with immunoblotting localized the VIM
FIG. 4. Immunofluorescence labeling with VIM 3B4 antibody on microinjected mouse skin fibroblasts (MSF) (a and b) and SW13 F-8 cells (c and d). Human protofilaments microinjected into an MSF cell and fixed after 15 min are randomly distributed throughout the cytoplasm (a). MSF cell (b) and SW13-F8 cell (c) transfected with the PVIM’~‘~~ vector and fixed after 20 h exhibit labeling of vimentin filaments; note the absence of labeling of endogenous vimentin in surrounding nontransfected MSF cells. Corresponding phase contrast picture to (c) is shown in (d). Magnification of a, c, and d, 800X; b, 500x.
3B4 epitope on the rod domain of vimentin. On the basis of sequence comparison , only an amino acid exchange at position 353 of the mouse vimentin rod domain was compatible with the recognition pattern of the VIM 3B4 antibody in immunocytochemical studies. At this position, valine in human [31, 321 and chicken  vimentin is replaced by leucine in rodent vimentin, suggesting that this conservative amino acid exchange caused the different reactions of VIM 3B4. Conversion of leucine at position 353 of the murine vimentin sequence to valine by in vitro mutagenesis of the murine cDNA followed by microinjection of a recombinant plasmid directing constitutive expression of the murine vimentinvd353 protein into MSF revealed staining with the VIM 3B4 antibody only in microinjected cells and thus verified the suggestion made above. In conclusion, the VIM 3B4 antibody can differentiate between vimentin molecules containing valine (positive reaction) and leucine (negative reaction) at amino acid residue 353 in the a-helical 2B part (i.e., carboxy-terminal proximal) of the central rod domain. Similarly, a single conservative amino acid substitution in the Cterminal end of the rod of gastropod IF proteins was found to be responsible for either positive or negative reaction with the IFA antibody . Previous data showed that VIM 3B4 reacts also with amphibian vimentin, which contains an isoleucine residue at the corresponding position . It indicates that isoleucine can substitute for valine, without impairing the reaction with VIM 3B4. This is compatible with the fact that valine and isoleucine both branch at the p-carbon, thus their conformational preferences are similar and differ from that of leucine . Furthermore, the average accessible area of valine and isoleucine in proteins is nearly identical and lower than that given for leucine [ 351. The rod domain of IF proteins, in general, consists of many imperfect repeats of seven amino acids, commonly designated a-g, where positions a and d are usually apolar and charged residues commonly occur at positions e and g. It is generally accepted that the rod domains of two IF monomers twist about each other in axial register in a coiled-coil arrangement, such that residues a and d of the parallel chains are brought together on the interior of the structure ([36-38,3]; discussed in ). Position 353 in both mouse and human vimentin corresponds to a “c” residue in the heptad repeat [5,40] and would be predicted to be located on the surface of the coiled-coil structure, being easily accessible to antibodies. This is also compatible with the suggestion that the location of epitopes recognized by antibodies generally correlates with the most exposed regions of a peptide surface . So far, amino acid sequences of rodent vimentin are only available for mouse and hamster. However, the immunocytochemical recognition pattern of VIM 3B4 strongly suggests that leucine in position 353 is not only
typical of mouse vimentin but may be unique for vimentin of all rodent species. Vimentin of the closely related order lagomorphes for instance, represented by the rabbit cell line Sfl EP in our study, exhibited a positive reaction with VIM 3B4. The use of species-specific vimentin antibodies in conjunction with an appropriate vimentin gene expression vector system has proven to be of great value for studies on vimentin assembly in transfected cells. The heterologous systems described hitherto were based on expression of either chicken vimentin in mouse cells  or mouse vimentin in human cells . Newly synthesized vimentin proteins were localized on the preexisting vimentin filament network, and the distribution suggested that vimentin subunits are randomly incorporated into the filaments at numerous sites throughout the network . However, possible aberrant interactions of related vimentin proteins as a result of speciesspecific differences in amino acid sequence could not completely be excluded and had to be discussed extensively. Differences between chicken and mammalian vimentin are most pronounced within the N-terminal head domain, showing only 60% amino acid identity, versus 93% identity in the rod domain . Particular emphasis has been placed on the arginine rich head domain of vimentin as a region influencing filament stability [43, 441 and representing the target of different kinases [45, 461 that probably modify filament subunit interaction and affect IF organization in uivo [47,48]. Expression of the murine vimentinv”‘353 protein in mouse cells represents a homologous system that rules out discussions on a possible influence of species-specific differences on vimentin assembly. Thus the VIM 3B4 antibody detection system, combined with the basic murine vimentinVa’353 expression system developed by us, may be a useful tool for further studies on the biology of IF proteins. This is demonstrated by the appearance of a well spread murine vimentin”1353 network in SW13-F8 and MSF cells which could be clearly distinguished from endogenous vimentin with the VIM 3B4 antibody. The results agree with those of others  obtained by transfection of SW13 cells, although the level of expression of the murine vimentinVaL353 protein in individual cells from the microinjected expression vector was much higher. In addition, human vimentin subunits introduced into living rodent cells could be clearly distinguished by the VIM 3B4 antibody from the preexisting filamentous vimentin network in these cells. A prior modification of the injected subunits, i.e., biotin-labeling [ 131 or fluorescence-labeling  along with attendant methological problems, is therefore unnecessary. The system established here should further permit an in uiuo analysis of phenotypes of various additional alterations in the murine vimentin’d353 protein that, of course, can be conveniently generated by in vitro mutagenesis or modification of the plasmid expression vec-
tors. Experiments employing the basic murine vimentinva1353 expression system and VIM 3B4 antibody detection system in MSF cells are currently being performed. The authors thank Dr. R. L. Shoeman (critical reading); Dr. E. de Groot (oligo synthesis); and U. Traub, M. Bialdiga, and U. Neumayer (cell culture). The Heinrich-Pette-Institut is financially supported by Freie und Hansestadt Hamburg and Bundesministerium fiir Gesundheit. REFERENCES 1.
2. 3. 4. 5.
6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16.
Franke, W. W., Schmid, E., Schiller, D. L., Winter, S., Jarasch, E. D., Moll, R., Denk, H., Jackson, B. W., and Illmensee, K. (1981) Cold Spring Harbor Symp. Qwnt. Biol. 46,431-459. Renner, W., Franke, W. W., Schmid, E., Geissler, N., Weber, K., Mandelkow, E. (1981) J. Mol. Biol. 149, 285-306. Aebi, U., Haner, J., Troncoso, J., Eichner, R., and Engel, A. (1988) Protoplasma 145, 73-81. Geisler, N., Kaufmann, E., Fischer, S., Plessmann, U., and Weber, K. (1983) EMBO J. 2, 1295-1302. Geisler, N., and Weber, K. (1986) in Cell and Molecular Biology of the Cytoskeleton @hay, J. W., Ed.), pp. 41-68, Plenum, New York/London. Osborn, M., and Weber, K. (1986) Trends Biol. Sci. 11,469-472. Lazarides, E. (1980) Nature 283, 249-256. Franke, W. W., Schmid, E., Osborn, M., and Weber, K. (1978) Proc. Natl. Acad. Sci. USA 75, 5034-5038. Zehner, Z. E., Li, Y., Roe, B. A., Paterson, B. M., and Sax, C. M. (1987) J. Biol. Chem. 262,8112-8120. Osborn, M., Altmannsberger, M., Debus, E., and Weber, K. (1984) Cold Spring Harbor Symp. Quant. Biol. 49, 191-200. Leader, M., Collins, M., Patel, J., and Henry, K. (1987) Histopathology 11,63-72. Domagala, W. M., Weber, K., and Osborn, M. (1988) Acta Cytol. 32,49-59. Vikstrom, K. L., Borisy, G. G., and Goldman, R. D. (1989) Proc. Natl. Acad. Sci. USA 86,549-553. Mittal, B., Sanger, J. M., and Sanger, J. W. (1989) Cell Motil. Cytoskel. 12, 127-138. Ngai, J., Coleman, T. R., and Lazarides, E. (1990) Cell 60,415427. Wiegers, W., Honer, B., and Traub, P. (1991) Cell Biol. Znt. Rep.
18. 19. 20. 21. 22.
Paulin-Levasseur, M., Scherbarth, A., Giese, G., Roser, K., Bohn, W., and Traub, P. (1989) J. Cell Sci. 92, 361-370. Hedberg, K. K., and Chen, L. B. (1989) Exp. Cell Res. 163,509517. Osborn, M., Debus, E., and Weher, K. (1984) Eur. J. Cell Biol. 34,137-143. Herrmann, H., Fouquet, B., and Franke, W. W. (1989) Deuelopment 105,279-298. Giese, G., and Traub, P. (1986) Eur. J. Cell Biol. 40, 266-274. Hawkes, R., Niday, E., and Gordon, J. (1982) Anal. Biochem. 119, 142-147.
Received September 4, 1991 Revised version received February
Nelson, W. J., Vorgias, C. E., and Traub, Biophys. Res. Commun. 106,1141-1147.
P. (1982) B&hem.
24. Shoeman, R. L., Honer, B., Stoller, T. J., Kesselmeier,
C., Miedel, M. C., Traub, P., and Graves, M. C. (1990) Proc. Natl. Acad. Sci. USA 87,6336-6340.
Shoeman, R. L., Mothes, E., Kesselmeier, (1990) Cell Biol. Znt. Rep. 14, 583-594. Hennekes,
H., Kuhn, S., and Traub,
C., and Traub,
P. (1990) Mol. Gen. Genet.
Kunkel, T. A., Roberts, J. D., and Zabour, R. A. (1987) in Methods in Enzymology (Wu, R., and Grossman, L., Ed.), Vol. 154, pp. 367-382, Academic Press, San Diego, CA.
Capetanaki, Y., Kuisk, Oncogene 5,645-655.
Quax, W., Egberts, W. V., Hendriks, W., Quax-Jeuken, Bloemendal, H. (1983) Cell 35,215-223.
Ferrari, S., Battini, R., Kaczmarek, L., Rittling, S., Calabretta, B., de Riel, J. K., Philiponis, V., Wei, J.-F., and Baserga, R. (1986) Mol. Cell. Biol. 6, 3614-3620.
Perreau, J., Lilienbaum, Gene 62, 7-16.
A., Vasseur, M., and Paulin,
Riemer, D., Dodemont, Biol. 56. 351-357.
H., and Weber, K. (1991) Eur. J. Cell
Richardson, J. S., and Richardson, D. C. (1989) in Prediction of Protein Structure and the Principles of Protein Conformation (Fasman, G. D., Ed.), Plenum, New York/London.
Rose, G. D., and Dworkin, J. E. (1989) in Prediction Structure and the Principles of Protein Conformation G. D., Ed.), pp. 625-633, Plenum, New York/London.
T. A. (1985) Proc. Natl. Acad. Sci. USA 82,488-492.
K., and Starnes, S. (1990) Y., and
of Protein (Fasman,
J. F., and Perry, D. A. D. (1988) Znt. J. Biol. Macromol.
Steinert, P. M., and Roop, D. R. (1988) Annu. Rev. Biochem. 57, 593-625.
Weber, K., and Geisler, Qua&. Biol. 49,153-159.
Mothes, E., Shoeman, R. L., andTraub,
N. (1984) Cold Spring
P. (1991) J. Struct. Biol.
Quax, W., and Bloemendal, H. (1986) in Cell and Molecular Biology of the Cytoskeleton @hay, J. W., Ed.), pp. 109-130, Plenum, New York/London. Van Regenmortel, M. H. V. (1989) Zmmunol. Today 10, 266272. Sarria, A. J., Nordeen,
S. K., and Evans, R. M. (1990) J. Cell Biol.
111,553-565. 43. 44. 45. 46. 47. 48.
Traub, P., and Vorgias, C. E. (1983) J. Cell Sci. 63,43-67. Geisler, N., Kaufmann, E., and Weber, K. (1982) Cell 30,277286. Evans, R. M. (1988) FEBS Lett. 234.73-78. Ando, S., Tanabe, K., Gonda, C., and Inagaki, M. (1989) Biochemistry 28, 2974-2979. Geisler, N., and Weber, K. (1988) EMBO J. 7, 15-20. Chou, Y.-H., Rosevear, E., and Goldmann, Natl. Acad. Sci. USA 86, 1885-1889.
R. D. (1989) Proc.