Autoimmunity, 1991, Vol. 9, pp. 99-107 Reprints available directly from the publisher Photocopying permitted by license only
0 1991 Harwood
Academic Publishers GmbH Printed in the United Kingdom
TRANSLOCATION OF THE NUCLEAR AUTOANTIGEN LA TO CELL SURFACE: ASSEMBLY AND DISASSEMBLY WITH THE EXTRACELLULAR MATRIX
Autoimmunity Downloaded from informahealthcare.com by University of Newcastle on 01/01/15 For personal use only.
M. BACHMANN$, S. CHANG, A. BERND*, W. MAYET’, K.-H. MEYER ZUM BUSCHENFELDE’ and W.E.G. MULLER Johannes Gutenberg University of Mainz. Inst. Phy.rio1. Chem. Dueshergweg 6 6500 Mainz. F.R.G. and * University of Frankfurt, Dermarologv, Theodor-Stern-Kai, 6000 Frankfurt, F.R.G. ‘University of Mainz, I. Med. Clinic, LangenheckstraJOe,6500 Mainz, F.R.G. (Received August 31. 1990: infinal form November 22, 1990)
La (SS-B) protein is known as one major antigenic target for autoantibodies from patients with certain autoimmune diseases such as Sjoegrens’ syndrome or Lupus Erythematosus. La protein belongs to the so called “extractable nuclear antigens”. Here we report that La antigen is not restricted to the nucleus as one might deduce from the exclusive nuclear staining pattern of patient anti-La antibodies but after stimulation of serum-starved cells with 10 YOfetal calf serum (FCS) appears and stays for at least 45 min at the outer surface of CV-I cells being available for binding of anti-La antibodies. In addition we found that a minor part of La antigen associates with the extracellular fibronectin network. After addition of 10 YOFCS to serum starved cells this extracellular autoantigen disassembled from the extracellular matrix and was taken up again by the cells. Incubation of serum starved cells with mercuric chloride, a known potent inducer of autoantibodies, also resulted in a detachment of the extracellular matrix associated La protein. From our studies it becomes likely that La protein itself is the antigen during autoimmunization. Moreover, once developed, anti-La antibodies might be able to bind to cell surface expressed La protein resulting in a damage of these cells leading to the inflammational events known to occur during disease.
KEY WORDS: Autoimmunity, La antigen, fibronectin.
inflammation on the cell surface and autoantibodies circulating in the peripheral blood bind to these cell surface expressed antigens resulting in a local cell damage and inflammation. At least in the case of the ANAs this seems unlikely. Based on their nuclear staining pattern in immunofluorescence microscopy the target antigens of ANAs, known as the Sm, RNP, Ro and La antigens, are deduced to localize exclusively in the n u c l e u ~ However, ~~~. during the last years it became evident that the exclusive nuclear staining pattern of these ANAs does not fit with structural and functional biochemical approaches. One well established example are the uridine rich small nuclear ribonucleoprotein particles (U snRNPs). They assemble in the cytoplasm4. For this purpose the newly transcribed U snRNAs are transported to the cytoplasm4.There they associate with a set of proteins among them the Sm and RNP antigens as a prereqisite for the relocation of the functional U snRNPs into the nucleus. For this purpose up to 70% of the total amount of the Sm proteins are stored in the Consequently, the reason for the lack of staining of this cytoplasmic part of the autoantigens with patient antibodies is unclear. Another example is the autoantigen La. During last years we studied the structure and function of this p r ~ t e i n ’ ” ~Using . a monoclonal anti-La mab we established a nucleo-
INTRODUCTION There is currently much interest in the elucidation of the molecular mechanism of autoimmunity and inflammation. Circulating autoantibodies have been described for many different human autoimmune diseases; the significance of their presence and their involvement in disease is a subject for debate’. Especially in the case of the anti-nuclear antibodies (ANAs) the hypothesis that the inflammation processes are only caused by circulating immunecomplexes formed as an result of damaged cells does not sufficiently explain (i) the reason for the first damaged cell (ii) why patients with the same set of antibodies develope different forms of disease and (iii) why inflammation is not permanent though the autoantibodies are present both during active disease and remission and (iv) why about 2 % of healthy blood donors contain autoantibodies never developing disease and inflammation. A simple explanation would be that the autoantigens are presented prior to GTo whom correspondence should be addressed. Dedicated to Prof. H . Holzmann (Dermatology - Frankfurt) on occasion of his 60th birthday. Correspondence address: Privat-Dozent Dr. Michael Bachmann, Johannes Gutenberg University of Mainz, Inst. Physiol. Chem. Duesbergweg 6 , 6500 Mainz, F.R.G. 99
100
M. BACHMANN E T A L .
Autoimmunity Downloaded from informahealthcare.com by University of Newcastle on 01/01/15 For personal use only.
cytoplasmic shuttling of La proteiri both, under physiological and pathophysiological conditions, including virus infections and UV-irradiati~n'-'~,'~-'' A detection of nuclear antigens on the cell surface of UV irradiated keratinocytes were also found with autoimmune patient sera". Moreover, recently Baboonian et al. reported a translocation of La protein in adenovirus infected cells". Similar shuttling processes were reported by us for herpes virus infected Here we describe simply reproducible conditions allowing us to translocate a La epitope onto the cell surface of epithelial cells and an assembly and disassembly with the extracellular matrix. Outside of the cells and on the cell surface of viable cells the La epitope was available for antibody binding. Therefore, we discuss that La protein might be directly involved in autoimmune reactions. MATERIALS AND METHODS Cell culture and immunolabeling
Sephrarose 4B and used as an immunoadsorbend for patient anti-La antibodies. Biotin labeling and isolation of labeled membrane proteins were performed as described by Hunter and Phillips". Briefly, membrane proteins of living serum stimulated cells were biotinylated. The biotinylated membrane proteins were isolated by affinity chromatography using Streptavidin covalently linked to Sepharose 4B. Polyacrylamide Gelelectrophoresis and immunoblotting of isolated membrane proteins was performed as described earlier". Epifluorescence microscopy
Stained cells were analyzed with a Zeiss Axiophot microscope equipped with epifluorescence optics and filter systems specific for the used fluorescence dyes'-''. Photographs were taken using Kodak 400 Tri-X Pan film or Kodak Ektachrome 400. Confocal laser scanning microscopy
Confocal laser scanning microscopy was performed CV- 1 cells (ATCC CCL 70) were grown to confluency using the viable labeled cells (see above). The used on coverslips in a humified CO, atmosphere in RPMI LSM 10 (Carl Zeiss, Oberkochen, F.R.G.) automatic1640 medium containing 1 0 % FCS. Then the cells ally generated 20 horizontal optical section images were starved into Go- phase". For this purpose the (5 122 pixels/8 bit, objective lenses Plan-Neofluar 40 x/ 10 OO/ serum/medium was replaced by 0.5 YO serum/ 1.3 oil, laser line Ar 488, emission filters BP 515-565) medium and the cells were starved for at least 2d. at z-distance of 200 nm through the entire cells Subsequently the 0.5 YO serum/medium was again (maximum thickness (2-4 pm)). Evaluation of the exchanged. Thereby the serum concentration was stored stacks of the horizontal optical sections was raised up again to 10%. The cells were further performed with the LSM 10 image processing unit. incubated for 1 h. 5 min prior to fixation of cell culture supernatant of anti-La mabs producing cells were Microinjection added. Then the cells were fixed with methanol, containing 0.02 YO ethylene glycol-bis (fi-aminoethyl Isolated homogenously purified La protein" was ether) N,N,N',N'-tetraacetic acid at - 20°C for 1 h. directly labeled with FITC and microinjected into Cell surface bound anti-La mab was detected by CV-I cells as described earlierI4. incubation of the cells with a species specific antimouse Ig-FITC conjugate (Medac, Hamburg; F.R.G.; dilution 1 : IOOO) as d e s ~ r i b e d ~ .~For ~ ~ the ~ ~ " 'RESULTS ~ double immunolabeling experiment the cells were first Exposure of a La epitope on the cell surface of stained for the La epitope as described before. Then growth stimulated CV-I cells the cells were incubated with the IgG fraction of a polyclonal anti-serum directed against fibr~nectin'~. From earlier studies it was evident that La protein The serum was developed in rabbit. The IgG fraction exists in both compartments the nucleus and the was directly labeled with rhodamin-isothiocyanat c y t o p l a ~ r n ~ -under ' ~ , culture conditions optimal for (RITC). This technique was successfully used in cells to proliferate, i.e. if they were incubated in earlier s t ~ d i e s ~ - l ~ ~ ' " - ' ~ . medium supplemented with a serum concentration of Monospecific patient anti-La antibodies were either greater 5 YO.The original impetus of out presented prepared by immunoaffinity chromatography to hom- data was to verify earlier studies on the shuttling of La ogenously purified La protein as described and suc- protein using now synchronized cells. For this cessfully used in earlier studie~~-"."-'~ or a generous purpose CV-I cells were exposed first to a serum gift of Prof. Dr. W. van Venroij (Nijmegen, The starvation block to arrest the cells in the Gocell cycle Netherlands). In the latter case a recombinant La phase" and then stimulated with 10 Yo FCS (fetal calf protein purified by the use of the anti-La mab La1 lG7 serum). During these studies we observed a cell from the bacteria extract was covalently linked to surface staining of viable cells with the anti-La rnab".
Autoimmunity Downloaded from informahealthcare.com by University of Newcastle on 01/01/15 For personal use only.
TRANSLOCATION OF LA AUTOANTIGEN
101
Figure 1 Immunolabeling for La antigen in serum-starved CV-1 cells, activated for I h with 10% FCS analyzed with conventional epifluorescence microscopy. CV-I cells were starved for 2 d in medium containing 0.5% FCS. Then the serum concentration was raised to 10%. Labeling was performed as described under methods. (A) serum stimulated cell, (B) correspondend phase contrast image; (C) non-stimulated cells, (D) correspondend phase contrast image.
As shown in Figure 1 after a stimulation of serum starved cells with 10 YO FCS the living stained cells give a punctuated cell surface staining (Figure 1A and B). In contrast, non-stimulated serum starved cells fail to give a similar staining pattern (Figure 1C and D). In addition, removing FCS from serum stimulated cells the punctuated cell surface staining pattern disappeared during the following 45 min (not shown). For unknown reasons the punctuated cell surface staining briefly reappeared without a further stimulation about 2 h after the first growth stimulation step (not shown). The cell surface labeling was highly specific for the anti-La mab. We incubated under identical conditions viable cells with a series of other mabs including antibody directed to the Sm, the RNP and several other monoclonal anti-U snRNP antibodies (not shown), including the anti-trimethyl cap antibody. None of these antibodies gave a similar
staining pattern (not shown). On the other hand a most recently characterized anti-La mab (La1 1G7)" gave identical results (not shown). In order to ensure that the immune reaction occurred indeed at the cell surface two additional approaches were performed: (i) membrane proteins of serum stimulated cells were biotinylated and isolated according to Hunter and Phillips" (see also Materials and methods) and (ii) we analyzed viable stained cells by confocal laser scanning immunofluorescence microscopical technique (CLSM). (i) Isolated biotinylated membrane proteins were separated by Polyacrylamide Gelelectrophoresis and transferred to PVDF-membrane. As shown in Figure 2 monospecific patient anti-La antibodies either adsorbed to homogenously purified La protein (lane a) or adsorbed to a recombinant La protein preparation (lane b) recognized a membrane protein with a
M. BACHMANN ET AL.
102
95 66
a b c
-”
L5-
I
31 Autoimmunity Downloaded from informahealthcare.com by University of Newcastle on 01/01/15 For personal use only.
To rule out that the cell surface is an artefact due to a leakage of nuclear La protein during our fixation procedure we microinjected fluorescently labeled La protein into the nuclei or cytoplasm of serum starved cells. As shown in Figure 4A and B the staining pattern of microinjected cells did not alter during our fixation process, especially there was no leakage out of the nucleus.
Figure 2 To support that La protein is expressed on the cell surface we labeled cell surface proteins with biotin. The biotin labeled membrane proteins were affinity purified and blotted against himunoadsorbed monospecific anti-La antibodies adsorbed to recombinant La preparation (lane b) or control human antibodies (lane c).
molecular weight of around 50-kD, the molecular weight of La protein. (ii) Using CLSM cells were cut into 200 nm sections starting at the upper site of the cellular membrane from the bottom to the top of the cells. As shown in Figure 3 (a to p) the punctuated stained regions follow the cell surface.
Releuse o j La epitope into the extrucellulur space
Another interesting observation was obtained performing our control studies. As shown in Figure 1 (C and D) the La epitope was not detectable on the cell surface of serum starved cells. However, during this analysis a staining of the extracellular matrix was seen. The staining of the extracellular matrix seemed to be non-randomly distributed. Mostly the material was seen near regions where cells started to divide. However, this observation needs further investigations. Nevertheless, a comparison of the image from immunolabeled serum starved cells with the corresponding phase contrast micrograph (Figure 5A to D) shows that the La epitope is associated with an extracellular filament. Applying the same immunocytochemical approach, using polyclonal antibodies raised
Figure 3 lmmunolabeling fcr La antigen in serum-starved CV-1 cells, activated for 1 h with 10% FCS analyzed with confocal laser scanning microscopy (CLSM). The distance between every section (a to p) IS 200 nm.
Autoimmunity Downloaded from informahealthcare.com by University of Newcastle on 01/01/15 For personal use only.
TRANSLOCATION O F LA AUTOANTIGEN
103
Figure 4 Microinjection of fluorescently labeled La protein into CV-I cells. To rule out that the cytoplasmic o r the cell surface labeling is an artefact due to the high solubility of La protein, we microinjected fluorescently labeled La protein into CV-I cells. Comparing the staining pattern of living cells (A) with the same cells after fixation (B) it is evident that our fixation procedure does not alter the intracellular localization.
Figure 5 Extracellular localization of La protein in quiescent CV- I cells. Carefully analyzing non-stimulated cells immunolabeled as shown in Figure I (C and D) a minor part of the La epitope was found extracellularely assembled with a fibrillar network (A). From the phase contrast images (B to D) taken in different planes it is evident that the labeled fiber is (i) outside of the cells and, (11) interconnects two cells.
Autoimmunity Downloaded from informahealthcare.com by University of Newcastle on 01/01/15 For personal use only.
104
M . B A C H M A N N ET A L
Figure 6 Assembly of the La epitopc with fibronectin. The cxtraccllular filament labeled with the anti-La mab ( A ) were costained with antibodies to fibronectin (B). Though the staining pattern is not completely identical it is evident that the extracellular filament system represents the fibronectin network.
against fibronectin2', these filaments could be identified as fibronectin (Figure 6A and B). Next we had to elucidate if the extracellular, fibronectin-associated La epitope disappears after serum activation of the Go-phase-arrested cells. The experiments revealed that after addition of 10 % FCS to the quiescent cells the La epitope disassembled from the extracellular matrix in dependence on time (Figure 7). The experiments showed that after a 30min incubation period the size of the fibrillar network alters and shows now a drop-like appearance (Figure 7A and B). Around 45 min after serum stimulation and in most cases still before the appearance of the La epitope on the cell surface most of the extracellular matrix associated La protein is taken up from the cells (Figure 7C and D). Destruction of the e.utracellular La network by mercuric chloride
Mercuric chloride (HgCI,) is known to induce the production of autoantibodies in experimental animals" by an unknown mechanism. We asked the question if HgCI, can mimic this effect also in vitro. In this first approach we determined the influence of HgC&on the integrity of the filamentous extracellular
system, shown to be associated with the La epitope. Therefore, we incubated the serum starved cells with HgCI, at a concentration (1 pg/ml) which is not cytotoxic for CV-1 cells and other cell types in vitro". As shown in Figure 7 (E and F) mercuric chloride was able to displace the La epitope from the extracellular matrix following a similar mechanism as naturally occuring after serum stimulation. DISCUSSION
In contrast to the culture conditions used in earlier studies to localize La antigen we used noncontinuously proliferating CV-1 cells in the present study. The cells were first grown in culture medium containing low amounts of serum (0.5 % of FCS) to enrich the cells in G,-G, phase of the cycle (19); then they were treated with fresh 10 % FCS. The rationale of this schedule was deduced from the experimental findings which indicate that the kinetics of cell cycle progression of cells obtained from patients with autoimmune diseases appears to be altered at least under in vitro conditions. E.g. it has been shown previously that P-lymphocytes from patients with Systemic Lupus Erythematosus (SLE) show ex vivo a
105
Autoimmunity Downloaded from informahealthcare.com by University of Newcastle on 01/01/15 For personal use only.
TRANSLOCATION OF LA AUTOANTIGEN
Figure 7 Disassembly of the La epitope from the extracellular matrix after serum stimulation (A to D) or HgCI, treatment ( E and F). After stimulation of serum starved CV-I cells with FCS the continuous staining of the fiber alters and drop like structures appear (A and C). The drop like structures disappear during the following 45 rnin (A). 15 min after serum stimulation, (C) 30 min after growth stimulation. Consequently. the La epitope is either uptaken from the cells or extracellularly desintegrated. Similarly, a treatment of quiescent non-stimulated cells (E) and (F) with HgClz also results in a release of the La epitope from the extracellular matrix following a similar mechanism as after growth stimulation (E) fluorescence image and (F) corresponding phase contrast image.
Autoimmunity Downloaded from informahealthcare.com by University of Newcastle on 01/01/15 For personal use only.
106
M . BACHMANN ET A L
disturbed proliferation b e h a ~ i o u ror ~ ~ it has been reported that sera from some SLE patients contain non-complement-dependent antibodies that inhibit cell proliferation*’. Therefore, we studied if growthmodulating factors present in FCS (10 %)” alter the localization of the La antigen in serum starved cells. The first novel result of this study is the demonstration that the autoantigen La translocalizes to the cell surface if serum starved cells were activated by serum growth factors and that anti-La antibodies were able to bind to such treated viable cells. Though it is unimportant for a direct involvement of autoantibodies in disease, especially in inflammation, if the cell surface protein carrying the antigenic epitope is really the La protein or shares only a common epitope with the autoantigen, it seems likely that La protein itself is translocalized to cell surface for the following reasons: (i) in earlier attempts the used anti-La mab was carefully characterized and shown not to cross react (ii) the with any additional component in CV-1 staining pattern was obtained on non-fixed, viable cells (iii) binding of the mab occurred also at a temperature of 4°C (data not shown) (iv) the staining disappeared from the cell surface if the growth stimulating factors were removed by washing and the cells were further incubated in the absence of growth factors (v) starved cells lack the cell surface staining (vi) the cell surface staining pattern was specific for anti-La mabs while other mabs did not give a similar staining pattern incubated under identical conditions, (vii) the punctuated pattern follow the cell surface as analyzed with the CLSM-technique, (viii) a protein with a molecular weight of La protein can be isolated from membrane fractions and (ix) the isolated membrane protein is recognized by immunoadsorbed patient anti-La antibodies, including anti-La antibodies adsorbed to a recombinant La protein. The second interesting finding towards an elucidation for the molecular events of inflammation is the documentation that serum-starved cells deposit a part of the autoantigen on the extracellular fibronectin network. This multifunctional molecule is known to comprise a series of specific binding sites*’, among them also nucleic acid/anti-nucleic acid complexes from SLE patients”. If the serum-starved cells are stimulated again with 1 0 % FCS, the La protein disappears from the extracellular space and appears on the cell surface, as reported above. At present we assume that the La protein is taken up by the cells in response to the function of the growth factors; however, an extracellular degradation by a neutral serine p r o t e a ~ e *can ~ not be excluded to date. The conclusion from this part of the study is the strong suggestion that the extracellularly deposited La protein together with fibronectin (or fragments derived from it) modulates T lymphocyte activity in a manner suggested earlier”.
In the third part of this study the molecular basis of HgC1,-induced autoimmunity was investigated in vifro, Hithereto this effect of HgCI, was well established in in vivo systems”. It was described that during the HgC1,-induced autoimmunity in rats immune complexes are deposited in a linear pattern along the basement membrane3’. The in vitro data presented here confirm with these results. In addition we observed that HgCI, causes an rearrangement of the fibrillarely organized autoantigen which includes a detachment of the antigen from the extracellular network. Released from the extracellular matrix under non-physiological conditions the antigen could be an ideal target for immune competent cells to produce autoantibodies. Moreover it is tempting to speculate that the postulated La/anti-La complexes are deposited again at the fibronectin molecules in the basement membrane as proposed3’. . Taken together the presented data show that in contrast to the exclusive staining pattern of patient anti-La sera the autoantigen La can exist in the cytoplasm, on the cell surface and even outside of the cells. Acknowledgements
This investigation was supported by a grant from the Deutsche Forschungsgemeinschaft (Mu 348/7-7). We thank Prof. Dr. Walter van Venrooij (University of Nijmegen, The Netherlands) for the gift of the recombinant La protein. References I. 2. 3. 4.
5.
6.
7.
8.
9.
Phillips T M . Immune complex assays: diagnostic and clinical application. Crit Rev Clin Lab Sci, 1989: 27; 237-264 Hardin JA, Mimori T. Autoantibodies to ribonucleoproteins. Clin Rheum Dis, 1986: 11; 485-505 Gale A, McCarty MD. Autoantibodies and their relation to rheumatic diseases. Advances in Rheumatol, 1986: 70; 237-261 Mattaj 1W. UsnRNP Assembly and Transport. In: Birnstiel, M.L., ed. Structure and Function of Major and Minor Small Ribonucleoprotein Particles. Springer-Verlag Berlin, Heidelberg, New York, Paris, Tokyo: p100. Sauterer RA, Feney RJ, Zieve GW. Cytoplasmic assembly of snRNP particles from stored protein and newly transcribed snRNA’s in L929 mouse fibroblasts. Exp Cell Res 1988: 176; 344-347 Sauterer RA, Goyal A, Zieve GW. Cytoplasmic assembly of small nuclear ribonucleoprotein particles from 6 S and 20 S RNA-free intermediates in L929 mouse fibroblasts. J Biol Chem, 1990: 265; 1048-1055 Bachmann M, Mayet WJ, Schroder HC, Pfeifer K, Meyer zum Biischenfelde KH, Miiller WEG. Association of La and Ro antigens with intracellular structures in HEp-2 carcinoma cells. Pror Natl Arad Sri U S A 1986: 83; 7770-7774 Bachmann M, Schroder HC, Falke D, Miiller WEG. Alteration of the intracellular localization of the La protein compared with the localization of U snRNPs. Cell Biol in1 Rep 1988: 12; 765-789 Bachmann M, Falke D, Schriider HC, Miiller WEG. IntracelM a r distribution of the La antigen in CV-1 cells after herpes
TRANSLOCATION O F LA AUTOANTIGEN
10.
11.
12.
Autoimmunity Downloaded from informahealthcare.com by University of Newcastle on 01/01/15 For personal use only.
13.
14. 15.
16.
17.
18.
19. 20.
simplex virus type I infection compared with the localization of U small nuclear ribonucleoprotein particles. J gen Virol 1989: 70; 881-891 Bachmann M, Pfeifer K, Schriider HC. Miiller WEG. The La antigen shuttles between the nucleus and the cytoplasm in cv-I cells. M o k c Cell Biochem 1989: 85; 103-1 14 Bachmann M, Pfeifer K, Schroder HC, Miiller WEG. Characterization of the autoantigen La as a nucleic-acid dependent ATPdse/dATPase with melting properties. Cell 1990: 60:8593 Carmo-Fonsecd M. Pfeifer K, Schroder HC, Vaz MF, Fonseca JE. Miiller WEG. Bachmann M. Identification of La ribonucleoproteins as a component of interchromatin granules. E s p Cell Res 1985; 185; 73-85 Bachmann M, Chang S, Slor H, Kukulies J, Miiller WEG. Shuttling of the autoantigen La between nucleus and cell surface after UV irradiation of human keratinocytes. Exp Cell Res. 1990; in press. Schroder HC, Bachmann M, Muller WEG. Methods for investigating nucleocytoplasmic transport. Fischer-Verlag. 1989. Bachmann M. Falke D, Miiller WEG. Is La protein involved in autoimmunization and inflammatory events during disease? Characterization of La protein as an unwinding enzyme. Mol Biol Rep, 1990: 14; 49-50 LeFeber WP, Norris DA, Ryan SR, Huff JC, Lee LA, Kubo M, Boyce ST, Kotzin BL, Weston WL. UV-light induces binding of antibodies selected nuclear antigens on cultured human keratinocytes. J Clin fnvrsr, 1984: 74; 1545-1551 Baboonian C, Venables PJW, Booth J, Williams DG, Roffe LM, Maini RN. Virus infection induces redistribution and membrane localization of the nuclear antigen La (SS-B): a possible mechanism for autoimmunity. Chi Exp fniniunol 1989: 78; 4 5 4 4 5 9 Springett G M , Moen RC, Anderson S, Blaese RM, Anderson WF. Infection efficiency of T lymphocytes with amphotropic retroviral vectors is cell cycle dependent. J Virol. 1989: 63; 3865-3869 Hunter A, Phillips JH. The recycling of a secretory granule membrane protein. Exp Cell Res, 1989: 182; 445-460 Gramzow M, Bachmann M, Uhlenbruck G , Dorn A, Miiller WEG. Identification and further characterization of the specific cell binding fragment from sponge aggregation factor. J Cell B i d , 1986: 102, 1344-1349
21
22.
23. 24. 25.
26. 27. 28.
29.
30.
31. 32.
107
Conrad J, Diehl-Seifert B, Zahn RK, Uhlenbruck G, Zimmermann E, Miiller WEG. Fibronectin is apparently not involved in species-specific reaggregation of cells from the sponge Geodia cydonium. J Cell Biochem, 1982: 19; 395-404 Reuter R, Tessars G . Vohr HW, Gleichman E, Liihrmann R. Mercuric chloride induces autoantibodies against U3 small nuclear ribonucleoprotein in susceptible mice. Proc Natl Acad Sci USA 1983: 86; 237-241 Christensen M, Mogensen SC, Rungby J. Toxicity and ultrastructural localization of mercuric chloride in cultured murine macrophages. Arch Toxicol, 1988: 62;4 4 0 4 4 6 Suzuki N, Sakane T. Induction ofexcessive B cell proliferation and differentiation by an in vitro stimulus in culture in human systemic lupus erythematosus. J Clin Invest, 1989: 83; 937-944 Bailey FA, Lilly M, Bertoli LF, Ball GV. An antibody that inhibits in vitro bone marrow proliferation in a patient with systemic lupus erythematosus and aplastic anemia. Arthritis Rheum. 1989: 32; 901-905 Pardee AB. A restriction point for control of normal animal cell proliferation. Proc Natl Acad Sci USA 1974: 71: 12861290 Yamada KM. Cell surface interactions with extracellular materials. Ann Rev Biochem 1983: 52;761-799 Gupta RC, Simpson WA, Raghow R. Interaction of fibronectin with DNA/anti-DNA complexes from systemic lupus erythematosus: Role of activated complement C I in modulation of the interactions. Cliti fmmunol fmmunopa/hol 1988: 46; 368-38 I Hantai D, Rao JS, Kahler C, Festoff BW. Decrease in plasminogen activator correlates with synapse elimination during neonatal development of mouse skeletal muscle. Proc Nail Acad Sci USA 1989: 86;362-366 Simon MM, Prester M , Nerz G, Kramer MD, Fruth U . Release of biologically active fragments from human plasmafibronectin by murine T cell-specific proteinase 1 (TSP-I). Biol Chem Hoppe Seyler 1988: 369; 107- I 12 Knoflach P, Albini B, Weiser MM. Experimental immune complexe disease of the intestine. fmmune fnvesr, 1989 18; 473-483 Rostagno AA, Frangione B, Gold LI. Biochemical characterization of the fibronectin binding sites for IgG. J fmmunol 1989: 143; 3277-3282
0 1991 Harwood
Academic Publishers GmbH Printed in the United Kingdom
TRANSLOCATION OF THE NUCLEAR AUTOANTIGEN LA TO CELL SURFACE: ASSEMBLY AND DISASSEMBLY WITH THE EXTRACELLULAR MATRIX
Autoimmunity Downloaded from informahealthcare.com by University of Newcastle on 01/01/15 For personal use only.
M. BACHMANN$, S. CHANG, A. BERND*, W. MAYET’, K.-H. MEYER ZUM BUSCHENFELDE’ and W.E.G. MULLER Johannes Gutenberg University of Mainz. Inst. Phy.rio1. Chem. Dueshergweg 6 6500 Mainz. F.R.G. and * University of Frankfurt, Dermarologv, Theodor-Stern-Kai, 6000 Frankfurt, F.R.G. ‘University of Mainz, I. Med. Clinic, LangenheckstraJOe,6500 Mainz, F.R.G. (Received August 31. 1990: infinal form November 22, 1990)
La (SS-B) protein is known as one major antigenic target for autoantibodies from patients with certain autoimmune diseases such as Sjoegrens’ syndrome or Lupus Erythematosus. La protein belongs to the so called “extractable nuclear antigens”. Here we report that La antigen is not restricted to the nucleus as one might deduce from the exclusive nuclear staining pattern of patient anti-La antibodies but after stimulation of serum-starved cells with 10 YOfetal calf serum (FCS) appears and stays for at least 45 min at the outer surface of CV-I cells being available for binding of anti-La antibodies. In addition we found that a minor part of La antigen associates with the extracellular fibronectin network. After addition of 10 YOFCS to serum starved cells this extracellular autoantigen disassembled from the extracellular matrix and was taken up again by the cells. Incubation of serum starved cells with mercuric chloride, a known potent inducer of autoantibodies, also resulted in a detachment of the extracellular matrix associated La protein. From our studies it becomes likely that La protein itself is the antigen during autoimmunization. Moreover, once developed, anti-La antibodies might be able to bind to cell surface expressed La protein resulting in a damage of these cells leading to the inflammational events known to occur during disease.
KEY WORDS: Autoimmunity, La antigen, fibronectin.
inflammation on the cell surface and autoantibodies circulating in the peripheral blood bind to these cell surface expressed antigens resulting in a local cell damage and inflammation. At least in the case of the ANAs this seems unlikely. Based on their nuclear staining pattern in immunofluorescence microscopy the target antigens of ANAs, known as the Sm, RNP, Ro and La antigens, are deduced to localize exclusively in the n u c l e u ~ However, ~~~. during the last years it became evident that the exclusive nuclear staining pattern of these ANAs does not fit with structural and functional biochemical approaches. One well established example are the uridine rich small nuclear ribonucleoprotein particles (U snRNPs). They assemble in the cytoplasm4. For this purpose the newly transcribed U snRNAs are transported to the cytoplasm4.There they associate with a set of proteins among them the Sm and RNP antigens as a prereqisite for the relocation of the functional U snRNPs into the nucleus. For this purpose up to 70% of the total amount of the Sm proteins are stored in the Consequently, the reason for the lack of staining of this cytoplasmic part of the autoantigens with patient antibodies is unclear. Another example is the autoantigen La. During last years we studied the structure and function of this p r ~ t e i n ’ ” ~Using . a monoclonal anti-La mab we established a nucleo-
INTRODUCTION There is currently much interest in the elucidation of the molecular mechanism of autoimmunity and inflammation. Circulating autoantibodies have been described for many different human autoimmune diseases; the significance of their presence and their involvement in disease is a subject for debate’. Especially in the case of the anti-nuclear antibodies (ANAs) the hypothesis that the inflammation processes are only caused by circulating immunecomplexes formed as an result of damaged cells does not sufficiently explain (i) the reason for the first damaged cell (ii) why patients with the same set of antibodies develope different forms of disease and (iii) why inflammation is not permanent though the autoantibodies are present both during active disease and remission and (iv) why about 2 % of healthy blood donors contain autoantibodies never developing disease and inflammation. A simple explanation would be that the autoantigens are presented prior to GTo whom correspondence should be addressed. Dedicated to Prof. H . Holzmann (Dermatology - Frankfurt) on occasion of his 60th birthday. Correspondence address: Privat-Dozent Dr. Michael Bachmann, Johannes Gutenberg University of Mainz, Inst. Physiol. Chem. Duesbergweg 6 , 6500 Mainz, F.R.G. 99
100
M. BACHMANN E T A L .
Autoimmunity Downloaded from informahealthcare.com by University of Newcastle on 01/01/15 For personal use only.
cytoplasmic shuttling of La proteiri both, under physiological and pathophysiological conditions, including virus infections and UV-irradiati~n'-'~,'~-'' A detection of nuclear antigens on the cell surface of UV irradiated keratinocytes were also found with autoimmune patient sera". Moreover, recently Baboonian et al. reported a translocation of La protein in adenovirus infected cells". Similar shuttling processes were reported by us for herpes virus infected Here we describe simply reproducible conditions allowing us to translocate a La epitope onto the cell surface of epithelial cells and an assembly and disassembly with the extracellular matrix. Outside of the cells and on the cell surface of viable cells the La epitope was available for antibody binding. Therefore, we discuss that La protein might be directly involved in autoimmune reactions. MATERIALS AND METHODS Cell culture and immunolabeling
Sephrarose 4B and used as an immunoadsorbend for patient anti-La antibodies. Biotin labeling and isolation of labeled membrane proteins were performed as described by Hunter and Phillips". Briefly, membrane proteins of living serum stimulated cells were biotinylated. The biotinylated membrane proteins were isolated by affinity chromatography using Streptavidin covalently linked to Sepharose 4B. Polyacrylamide Gelelectrophoresis and immunoblotting of isolated membrane proteins was performed as described earlier". Epifluorescence microscopy
Stained cells were analyzed with a Zeiss Axiophot microscope equipped with epifluorescence optics and filter systems specific for the used fluorescence dyes'-''. Photographs were taken using Kodak 400 Tri-X Pan film or Kodak Ektachrome 400. Confocal laser scanning microscopy
Confocal laser scanning microscopy was performed CV- 1 cells (ATCC CCL 70) were grown to confluency using the viable labeled cells (see above). The used on coverslips in a humified CO, atmosphere in RPMI LSM 10 (Carl Zeiss, Oberkochen, F.R.G.) automatic1640 medium containing 1 0 % FCS. Then the cells ally generated 20 horizontal optical section images were starved into Go- phase". For this purpose the (5 122 pixels/8 bit, objective lenses Plan-Neofluar 40 x/ 10 OO/ serum/medium was replaced by 0.5 YO serum/ 1.3 oil, laser line Ar 488, emission filters BP 515-565) medium and the cells were starved for at least 2d. at z-distance of 200 nm through the entire cells Subsequently the 0.5 YO serum/medium was again (maximum thickness (2-4 pm)). Evaluation of the exchanged. Thereby the serum concentration was stored stacks of the horizontal optical sections was raised up again to 10%. The cells were further performed with the LSM 10 image processing unit. incubated for 1 h. 5 min prior to fixation of cell culture supernatant of anti-La mabs producing cells were Microinjection added. Then the cells were fixed with methanol, containing 0.02 YO ethylene glycol-bis (fi-aminoethyl Isolated homogenously purified La protein" was ether) N,N,N',N'-tetraacetic acid at - 20°C for 1 h. directly labeled with FITC and microinjected into Cell surface bound anti-La mab was detected by CV-I cells as described earlierI4. incubation of the cells with a species specific antimouse Ig-FITC conjugate (Medac, Hamburg; F.R.G.; dilution 1 : IOOO) as d e s ~ r i b e d ~ .~For ~ ~ the ~ ~ " 'RESULTS ~ double immunolabeling experiment the cells were first Exposure of a La epitope on the cell surface of stained for the La epitope as described before. Then growth stimulated CV-I cells the cells were incubated with the IgG fraction of a polyclonal anti-serum directed against fibr~nectin'~. From earlier studies it was evident that La protein The serum was developed in rabbit. The IgG fraction exists in both compartments the nucleus and the was directly labeled with rhodamin-isothiocyanat c y t o p l a ~ r n ~ -under ' ~ , culture conditions optimal for (RITC). This technique was successfully used in cells to proliferate, i.e. if they were incubated in earlier s t ~ d i e s ~ - l ~ ~ ' " - ' ~ . medium supplemented with a serum concentration of Monospecific patient anti-La antibodies were either greater 5 YO.The original impetus of out presented prepared by immunoaffinity chromatography to hom- data was to verify earlier studies on the shuttling of La ogenously purified La protein as described and suc- protein using now synchronized cells. For this cessfully used in earlier studie~~-"."-'~ or a generous purpose CV-I cells were exposed first to a serum gift of Prof. Dr. W. van Venroij (Nijmegen, The starvation block to arrest the cells in the Gocell cycle Netherlands). In the latter case a recombinant La phase" and then stimulated with 10 Yo FCS (fetal calf protein purified by the use of the anti-La mab La1 lG7 serum). During these studies we observed a cell from the bacteria extract was covalently linked to surface staining of viable cells with the anti-La rnab".
Autoimmunity Downloaded from informahealthcare.com by University of Newcastle on 01/01/15 For personal use only.
TRANSLOCATION OF LA AUTOANTIGEN
101
Figure 1 Immunolabeling for La antigen in serum-starved CV-1 cells, activated for I h with 10% FCS analyzed with conventional epifluorescence microscopy. CV-I cells were starved for 2 d in medium containing 0.5% FCS. Then the serum concentration was raised to 10%. Labeling was performed as described under methods. (A) serum stimulated cell, (B) correspondend phase contrast image; (C) non-stimulated cells, (D) correspondend phase contrast image.
As shown in Figure 1 after a stimulation of serum starved cells with 10 YO FCS the living stained cells give a punctuated cell surface staining (Figure 1A and B). In contrast, non-stimulated serum starved cells fail to give a similar staining pattern (Figure 1C and D). In addition, removing FCS from serum stimulated cells the punctuated cell surface staining pattern disappeared during the following 45 min (not shown). For unknown reasons the punctuated cell surface staining briefly reappeared without a further stimulation about 2 h after the first growth stimulation step (not shown). The cell surface labeling was highly specific for the anti-La mab. We incubated under identical conditions viable cells with a series of other mabs including antibody directed to the Sm, the RNP and several other monoclonal anti-U snRNP antibodies (not shown), including the anti-trimethyl cap antibody. None of these antibodies gave a similar
staining pattern (not shown). On the other hand a most recently characterized anti-La mab (La1 1G7)" gave identical results (not shown). In order to ensure that the immune reaction occurred indeed at the cell surface two additional approaches were performed: (i) membrane proteins of serum stimulated cells were biotinylated and isolated according to Hunter and Phillips" (see also Materials and methods) and (ii) we analyzed viable stained cells by confocal laser scanning immunofluorescence microscopical technique (CLSM). (i) Isolated biotinylated membrane proteins were separated by Polyacrylamide Gelelectrophoresis and transferred to PVDF-membrane. As shown in Figure 2 monospecific patient anti-La antibodies either adsorbed to homogenously purified La protein (lane a) or adsorbed to a recombinant La protein preparation (lane b) recognized a membrane protein with a
M. BACHMANN ET AL.
102
95 66
a b c
-”
L5-
I
31 Autoimmunity Downloaded from informahealthcare.com by University of Newcastle on 01/01/15 For personal use only.
To rule out that the cell surface is an artefact due to a leakage of nuclear La protein during our fixation procedure we microinjected fluorescently labeled La protein into the nuclei or cytoplasm of serum starved cells. As shown in Figure 4A and B the staining pattern of microinjected cells did not alter during our fixation process, especially there was no leakage out of the nucleus.
Figure 2 To support that La protein is expressed on the cell surface we labeled cell surface proteins with biotin. The biotin labeled membrane proteins were affinity purified and blotted against himunoadsorbed monospecific anti-La antibodies adsorbed to recombinant La preparation (lane b) or control human antibodies (lane c).
molecular weight of around 50-kD, the molecular weight of La protein. (ii) Using CLSM cells were cut into 200 nm sections starting at the upper site of the cellular membrane from the bottom to the top of the cells. As shown in Figure 3 (a to p) the punctuated stained regions follow the cell surface.
Releuse o j La epitope into the extrucellulur space
Another interesting observation was obtained performing our control studies. As shown in Figure 1 (C and D) the La epitope was not detectable on the cell surface of serum starved cells. However, during this analysis a staining of the extracellular matrix was seen. The staining of the extracellular matrix seemed to be non-randomly distributed. Mostly the material was seen near regions where cells started to divide. However, this observation needs further investigations. Nevertheless, a comparison of the image from immunolabeled serum starved cells with the corresponding phase contrast micrograph (Figure 5A to D) shows that the La epitope is associated with an extracellular filament. Applying the same immunocytochemical approach, using polyclonal antibodies raised
Figure 3 lmmunolabeling fcr La antigen in serum-starved CV-1 cells, activated for 1 h with 10% FCS analyzed with confocal laser scanning microscopy (CLSM). The distance between every section (a to p) IS 200 nm.
Autoimmunity Downloaded from informahealthcare.com by University of Newcastle on 01/01/15 For personal use only.
TRANSLOCATION O F LA AUTOANTIGEN
103
Figure 4 Microinjection of fluorescently labeled La protein into CV-I cells. To rule out that the cytoplasmic o r the cell surface labeling is an artefact due to the high solubility of La protein, we microinjected fluorescently labeled La protein into CV-I cells. Comparing the staining pattern of living cells (A) with the same cells after fixation (B) it is evident that our fixation procedure does not alter the intracellular localization.
Figure 5 Extracellular localization of La protein in quiescent CV- I cells. Carefully analyzing non-stimulated cells immunolabeled as shown in Figure I (C and D) a minor part of the La epitope was found extracellularely assembled with a fibrillar network (A). From the phase contrast images (B to D) taken in different planes it is evident that the labeled fiber is (i) outside of the cells and, (11) interconnects two cells.
Autoimmunity Downloaded from informahealthcare.com by University of Newcastle on 01/01/15 For personal use only.
104
M . B A C H M A N N ET A L
Figure 6 Assembly of the La epitopc with fibronectin. The cxtraccllular filament labeled with the anti-La mab ( A ) were costained with antibodies to fibronectin (B). Though the staining pattern is not completely identical it is evident that the extracellular filament system represents the fibronectin network.
against fibronectin2', these filaments could be identified as fibronectin (Figure 6A and B). Next we had to elucidate if the extracellular, fibronectin-associated La epitope disappears after serum activation of the Go-phase-arrested cells. The experiments revealed that after addition of 10 % FCS to the quiescent cells the La epitope disassembled from the extracellular matrix in dependence on time (Figure 7). The experiments showed that after a 30min incubation period the size of the fibrillar network alters and shows now a drop-like appearance (Figure 7A and B). Around 45 min after serum stimulation and in most cases still before the appearance of the La epitope on the cell surface most of the extracellular matrix associated La protein is taken up from the cells (Figure 7C and D). Destruction of the e.utracellular La network by mercuric chloride
Mercuric chloride (HgCI,) is known to induce the production of autoantibodies in experimental animals" by an unknown mechanism. We asked the question if HgCI, can mimic this effect also in vitro. In this first approach we determined the influence of HgC&on the integrity of the filamentous extracellular
system, shown to be associated with the La epitope. Therefore, we incubated the serum starved cells with HgCI, at a concentration (1 pg/ml) which is not cytotoxic for CV-1 cells and other cell types in vitro". As shown in Figure 7 (E and F) mercuric chloride was able to displace the La epitope from the extracellular matrix following a similar mechanism as naturally occuring after serum stimulation. DISCUSSION
In contrast to the culture conditions used in earlier studies to localize La antigen we used noncontinuously proliferating CV-1 cells in the present study. The cells were first grown in culture medium containing low amounts of serum (0.5 % of FCS) to enrich the cells in G,-G, phase of the cycle (19); then they were treated with fresh 10 % FCS. The rationale of this schedule was deduced from the experimental findings which indicate that the kinetics of cell cycle progression of cells obtained from patients with autoimmune diseases appears to be altered at least under in vitro conditions. E.g. it has been shown previously that P-lymphocytes from patients with Systemic Lupus Erythematosus (SLE) show ex vivo a
105
Autoimmunity Downloaded from informahealthcare.com by University of Newcastle on 01/01/15 For personal use only.
TRANSLOCATION OF LA AUTOANTIGEN
Figure 7 Disassembly of the La epitope from the extracellular matrix after serum stimulation (A to D) or HgCI, treatment ( E and F). After stimulation of serum starved CV-I cells with FCS the continuous staining of the fiber alters and drop like structures appear (A and C). The drop like structures disappear during the following 45 rnin (A). 15 min after serum stimulation, (C) 30 min after growth stimulation. Consequently. the La epitope is either uptaken from the cells or extracellularly desintegrated. Similarly, a treatment of quiescent non-stimulated cells (E) and (F) with HgClz also results in a release of the La epitope from the extracellular matrix following a similar mechanism as after growth stimulation (E) fluorescence image and (F) corresponding phase contrast image.
Autoimmunity Downloaded from informahealthcare.com by University of Newcastle on 01/01/15 For personal use only.
106
M . BACHMANN ET A L
disturbed proliferation b e h a ~ i o u ror ~ ~ it has been reported that sera from some SLE patients contain non-complement-dependent antibodies that inhibit cell proliferation*’. Therefore, we studied if growthmodulating factors present in FCS (10 %)” alter the localization of the La antigen in serum starved cells. The first novel result of this study is the demonstration that the autoantigen La translocalizes to the cell surface if serum starved cells were activated by serum growth factors and that anti-La antibodies were able to bind to such treated viable cells. Though it is unimportant for a direct involvement of autoantibodies in disease, especially in inflammation, if the cell surface protein carrying the antigenic epitope is really the La protein or shares only a common epitope with the autoantigen, it seems likely that La protein itself is translocalized to cell surface for the following reasons: (i) in earlier attempts the used anti-La mab was carefully characterized and shown not to cross react (ii) the with any additional component in CV-1 staining pattern was obtained on non-fixed, viable cells (iii) binding of the mab occurred also at a temperature of 4°C (data not shown) (iv) the staining disappeared from the cell surface if the growth stimulating factors were removed by washing and the cells were further incubated in the absence of growth factors (v) starved cells lack the cell surface staining (vi) the cell surface staining pattern was specific for anti-La mabs while other mabs did not give a similar staining pattern incubated under identical conditions, (vii) the punctuated pattern follow the cell surface as analyzed with the CLSM-technique, (viii) a protein with a molecular weight of La protein can be isolated from membrane fractions and (ix) the isolated membrane protein is recognized by immunoadsorbed patient anti-La antibodies, including anti-La antibodies adsorbed to a recombinant La protein. The second interesting finding towards an elucidation for the molecular events of inflammation is the documentation that serum-starved cells deposit a part of the autoantigen on the extracellular fibronectin network. This multifunctional molecule is known to comprise a series of specific binding sites*’, among them also nucleic acid/anti-nucleic acid complexes from SLE patients”. If the serum-starved cells are stimulated again with 1 0 % FCS, the La protein disappears from the extracellular space and appears on the cell surface, as reported above. At present we assume that the La protein is taken up by the cells in response to the function of the growth factors; however, an extracellular degradation by a neutral serine p r o t e a ~ e *can ~ not be excluded to date. The conclusion from this part of the study is the strong suggestion that the extracellularly deposited La protein together with fibronectin (or fragments derived from it) modulates T lymphocyte activity in a manner suggested earlier”.
In the third part of this study the molecular basis of HgC1,-induced autoimmunity was investigated in vifro, Hithereto this effect of HgCI, was well established in in vivo systems”. It was described that during the HgC1,-induced autoimmunity in rats immune complexes are deposited in a linear pattern along the basement membrane3’. The in vitro data presented here confirm with these results. In addition we observed that HgCI, causes an rearrangement of the fibrillarely organized autoantigen which includes a detachment of the antigen from the extracellular network. Released from the extracellular matrix under non-physiological conditions the antigen could be an ideal target for immune competent cells to produce autoantibodies. Moreover it is tempting to speculate that the postulated La/anti-La complexes are deposited again at the fibronectin molecules in the basement membrane as proposed3’. . Taken together the presented data show that in contrast to the exclusive staining pattern of patient anti-La sera the autoantigen La can exist in the cytoplasm, on the cell surface and even outside of the cells. Acknowledgements
This investigation was supported by a grant from the Deutsche Forschungsgemeinschaft (Mu 348/7-7). We thank Prof. Dr. Walter van Venrooij (University of Nijmegen, The Netherlands) for the gift of the recombinant La protein. References I. 2. 3. 4.
5.
6.
7.
8.
9.
Phillips T M . Immune complex assays: diagnostic and clinical application. Crit Rev Clin Lab Sci, 1989: 27; 237-264 Hardin JA, Mimori T. Autoantibodies to ribonucleoproteins. Clin Rheum Dis, 1986: 11; 485-505 Gale A, McCarty MD. Autoantibodies and their relation to rheumatic diseases. Advances in Rheumatol, 1986: 70; 237-261 Mattaj 1W. UsnRNP Assembly and Transport. In: Birnstiel, M.L., ed. Structure and Function of Major and Minor Small Ribonucleoprotein Particles. Springer-Verlag Berlin, Heidelberg, New York, Paris, Tokyo: p100. Sauterer RA, Feney RJ, Zieve GW. Cytoplasmic assembly of snRNP particles from stored protein and newly transcribed snRNA’s in L929 mouse fibroblasts. Exp Cell Res 1988: 176; 344-347 Sauterer RA, Goyal A, Zieve GW. Cytoplasmic assembly of small nuclear ribonucleoprotein particles from 6 S and 20 S RNA-free intermediates in L929 mouse fibroblasts. J Biol Chem, 1990: 265; 1048-1055 Bachmann M, Mayet WJ, Schroder HC, Pfeifer K, Meyer zum Biischenfelde KH, Miiller WEG. Association of La and Ro antigens with intracellular structures in HEp-2 carcinoma cells. Pror Natl Arad Sri U S A 1986: 83; 7770-7774 Bachmann M, Schroder HC, Falke D, Miiller WEG. Alteration of the intracellular localization of the La protein compared with the localization of U snRNPs. Cell Biol in1 Rep 1988: 12; 765-789 Bachmann M, Falke D, Schriider HC, Miiller WEG. IntracelM a r distribution of the La antigen in CV-1 cells after herpes
TRANSLOCATION O F LA AUTOANTIGEN
10.
11.
12.
Autoimmunity Downloaded from informahealthcare.com by University of Newcastle on 01/01/15 For personal use only.
13.
14. 15.
16.
17.
18.
19. 20.
simplex virus type I infection compared with the localization of U small nuclear ribonucleoprotein particles. J gen Virol 1989: 70; 881-891 Bachmann M, Pfeifer K, Schriider HC. Miiller WEG. The La antigen shuttles between the nucleus and the cytoplasm in cv-I cells. M o k c Cell Biochem 1989: 85; 103-1 14 Bachmann M, Pfeifer K, Schroder HC, Miiller WEG. Characterization of the autoantigen La as a nucleic-acid dependent ATPdse/dATPase with melting properties. Cell 1990: 60:8593 Carmo-Fonsecd M. Pfeifer K, Schroder HC, Vaz MF, Fonseca JE. Miiller WEG. Bachmann M. Identification of La ribonucleoproteins as a component of interchromatin granules. E s p Cell Res 1985; 185; 73-85 Bachmann M, Chang S, Slor H, Kukulies J, Miiller WEG. Shuttling of the autoantigen La between nucleus and cell surface after UV irradiation of human keratinocytes. Exp Cell Res. 1990; in press. Schroder HC, Bachmann M, Muller WEG. Methods for investigating nucleocytoplasmic transport. Fischer-Verlag. 1989. Bachmann M. Falke D, Miiller WEG. Is La protein involved in autoimmunization and inflammatory events during disease? Characterization of La protein as an unwinding enzyme. Mol Biol Rep, 1990: 14; 49-50 LeFeber WP, Norris DA, Ryan SR, Huff JC, Lee LA, Kubo M, Boyce ST, Kotzin BL, Weston WL. UV-light induces binding of antibodies selected nuclear antigens on cultured human keratinocytes. J Clin fnvrsr, 1984: 74; 1545-1551 Baboonian C, Venables PJW, Booth J, Williams DG, Roffe LM, Maini RN. Virus infection induces redistribution and membrane localization of the nuclear antigen La (SS-B): a possible mechanism for autoimmunity. Chi Exp fniniunol 1989: 78; 4 5 4 4 5 9 Springett G M , Moen RC, Anderson S, Blaese RM, Anderson WF. Infection efficiency of T lymphocytes with amphotropic retroviral vectors is cell cycle dependent. J Virol. 1989: 63; 3865-3869 Hunter A, Phillips JH. The recycling of a secretory granule membrane protein. Exp Cell Res, 1989: 182; 445-460 Gramzow M, Bachmann M, Uhlenbruck G , Dorn A, Miiller WEG. Identification and further characterization of the specific cell binding fragment from sponge aggregation factor. J Cell B i d , 1986: 102, 1344-1349
21
22.
23. 24. 25.
26. 27. 28.
29.
30.
31. 32.
107
Conrad J, Diehl-Seifert B, Zahn RK, Uhlenbruck G, Zimmermann E, Miiller WEG. Fibronectin is apparently not involved in species-specific reaggregation of cells from the sponge Geodia cydonium. J Cell Biochem, 1982: 19; 395-404 Reuter R, Tessars G . Vohr HW, Gleichman E, Liihrmann R. Mercuric chloride induces autoantibodies against U3 small nuclear ribonucleoprotein in susceptible mice. Proc Natl Acad Sci USA 1983: 86; 237-241 Christensen M, Mogensen SC, Rungby J. Toxicity and ultrastructural localization of mercuric chloride in cultured murine macrophages. Arch Toxicol, 1988: 62;4 4 0 4 4 6 Suzuki N, Sakane T. Induction ofexcessive B cell proliferation and differentiation by an in vitro stimulus in culture in human systemic lupus erythematosus. J Clin Invest, 1989: 83; 937-944 Bailey FA, Lilly M, Bertoli LF, Ball GV. An antibody that inhibits in vitro bone marrow proliferation in a patient with systemic lupus erythematosus and aplastic anemia. Arthritis Rheum. 1989: 32; 901-905 Pardee AB. A restriction point for control of normal animal cell proliferation. Proc Natl Acad Sci USA 1974: 71: 12861290 Yamada KM. Cell surface interactions with extracellular materials. Ann Rev Biochem 1983: 52;761-799 Gupta RC, Simpson WA, Raghow R. Interaction of fibronectin with DNA/anti-DNA complexes from systemic lupus erythematosus: Role of activated complement C I in modulation of the interactions. Cliti fmmunol fmmunopa/hol 1988: 46; 368-38 I Hantai D, Rao JS, Kahler C, Festoff BW. Decrease in plasminogen activator correlates with synapse elimination during neonatal development of mouse skeletal muscle. Proc Nail Acad Sci USA 1989: 86;362-366 Simon MM, Prester M , Nerz G, Kramer MD, Fruth U . Release of biologically active fragments from human plasmafibronectin by murine T cell-specific proteinase 1 (TSP-I). Biol Chem Hoppe Seyler 1988: 369; 107- I 12 Knoflach P, Albini B, Weiser MM. Experimental immune complexe disease of the intestine. fmmune fnvesr, 1989 18; 473-483 Rostagno AA, Frangione B, Gold LI. Biochemical characterization of the fibronectin binding sites for IgG. J fmmunol 1989: 143; 3277-3282
