Prosomes, ubiquitous ribonucleoprotein (RNP) particles of defined biochemical and morphological structure, first isolated as a subcomplex of the repressed globin mRNP in avian and mouse erythroblasts, were also found in the cytoplasm of other vertebrates associated with other mRNAs. Here we show that prosomes are also present in the cell nucleus and, furthermore, that the cytolocalization of specific prosomal peptides is a function of differentiation. Four monoclonal antibodies, raised against the duck prosomal proteins, p27K, p28K, p29K and p31K (K = 103Mr) react to variable degree with prosomes of chicken, mouse, and human cells. Immunocytochemical and biochemical analyses show that all four antigens are present in both the cytoplasm and the nucleus of avian erythroblasts and avian erythroblastosis virus (AEV)-transformed erythroleukaemic cells. Interestingly, the prosomes disappear in the course of the terminal differentiation of erythroblasts to mature erythrocytes. Although all the four prosomal antigens tested are present in both the nuclear and cytoplasmic compartments, slight differences in the immunofluorescent patterns indicate that each antigen may have a particular cytological distribution that varies in the course of differentiation.

In eukaryotic cells, cytoplasmic mRNAs occur always in association with proteins, forming messenger ribonucleoprotein (mRNP) complexes (Spirin et al. 1964; Spirin, 1969; Morel et al. 1971; Scherrer, 1973; Brawerman, 1974; Vincent et al. 1981). Messenger RNP proteins are bound to either polyribosomal mRNAs engaged in protein synthesis, or to ‘free’ mRNAs in the cytoplasm, forming (ribosome-) free mRNPs; the latter contain the repressed, inactive form of mRNA (Vincent et al. 1977, 1981; Civelli et al. 1976, 1980). Selective translation of mRNAs has been extensively studied in avian erythroblasts (Imaizumi-Scherrer et al. 1982): some medium abundant mRNAs, e.g. the mRNA coding for the poly(A)-binding protein (Maundrell et al. 1983; Akhayat et al. 1983), are fully repressed. However, other mRNAs including globin mRNAs exist simultaneously in both active and repressed forms (Imaizumi-Scherrer et al.1982). The repressed globin mRNPs have been identified in duck erythroblasts as a ribonucleoprotein complex sedimenting at 20 S, untranslatable in vitro but containing mRNAs active after total deproteinization (Civelli et al. 1980).

Recently, the prosomes, a class of ubiquitous RNP particles of defined morphology and biochemical composition have been characterized in several cell types and species, as subcomponents of the repressed mRNP complexes (Schmid et al. 1984; Martins de Sa et al. 1986). The prosome is a highly stable RNP complex with a characteristic raspberry-shaped morphology in the electron microscope. This particle has a sedimentation coefficient of 19 S corresponding to an 4/r of about 600 000, and is composed of 15% RNA and 8 5% proteins. The protein composition includes about 25 polypeptides with Mr ranging from 21 000 to 35 000 and pl ranging from 4 to 7, with an additional 56 000 47, peptide characteristic of avian species. Comparative biochemical studies show that 16 of these 25 protein components have identical pI values and molecular weights in duck and mouse, and human cells (Martins de Sa et al. 1986).

In one case a prosome-like particle has been clearly characterized as being built of about 20 copies of a single (p2lK) protein component linked to its RNA (Akhayat, 1986; Akhayat et al. 1987). It is not clear at present if all prosomes are made up of single identical polypeptides, or if some are composed of multiple copies of different peptide components.

Prosomes associated with globin mRNP from mouse and duck erythroblasts contain two low molecular weight RNAs 70—90 nucleotides long, which show similar fingerprints in the two species. The prosomes of HeLa cell mRNPs associated with a larger and more complex population of repressed mRNA, containing at least 12 RNA species of 50–150 nucleotides. These prosomal RNAs (pRNA) have the unique property of hybridizing to a wide variety of messenger RNA types (Martins de Sa et al. 1986). Furthermore, prosome-like particles have the capacity to inhibit in vitro protein synthesis (Akhayat, 1986; Akhayat et al. 1987); interestingly, viral mRNA is 10—50 times more sensitive to this effect than cellular mRNAs (Horsch et al. 1987).

Recently, we have produced monoclonal antibodies (MAbs) directed against four different proteins from duck prosomes dissociated from the 20 S globin mRNP. Using these antibodies, immunological comparisons in several species shows that during evolution, some of the antigenic epitopes of prosomal proteins have been highly conserved. We present here the first investigation of the cytolocalization of the prosomes, using these antibodies. The cytological distribution of prosomes in adult and embryonic avian erythroblasts during terminal differentiation, or in AEV-transformed erythroblasts, was investigated by immunofluorescence and immunoelectron microscopy. Our results show that all four prosomal antigens are localized not only on cytoplasmic structures, but also to various extents in the nucleus; their quantitative distribution varies with the stage of differentiation and, possibly, with the type of antigen.

Preparation of free mRNPs and prosomes

Free globin mRNPs from the post-polyribosomal supernatant of immature duck red blood cells were isolated by preparative zonal centrifugation as described in detail previously (Gander et al. 1973; Vincent et al. 1977). Isolation and purification of the prosomes were as reported previously (Schmid et al. 1984; Martins de Sa et al. 1986).

Cells and culture conditions

PAI myeloma cells (courtesy of Dr Stahli, Hoffmann La Roche, Basel) a non-immunoglobulin (IgG) producing cell line and the hybridoma cells were cultivated in RPMI 1640 medium, complemented by 15% foetal calf serum (Boehringer) and 2 mM-glutamine, 1 mM-sodium pyruvate, penicillin (100 units ml−1) and streptomycin (100 μgml−1), in the presence of 10% CO2.

AEV-transformed erythroblasts (LSCC HD3 cell line) were grown on Dulbecco’s modified Eagle’s medium supplemented with 8% foetal calf serum and 2% chicken serum. Normal erythroblasts were taken from the blood of anaemic ducks and embryonic chickens.

Monoclonal antibody production

Immunization

Six- or seven-week-old female Balb/c mice were immunized with 20 μg total duck prosomes mixed in Freund’s complete adjuvant (Gibco), by subcutaneous injection into the footpads and several dorsal sites. Two booster injections of 20 μg prosomes in Freund’s incomplete adjuvant were given subcutaneously in the dorsal sites at 3 and 6 weeks after the first injection. Ten days after the second injection, a serum sample was taken from the tail vein to test for circulating antibody. After a further period of 2 months following the third injection, 50 μg of the antigens in phosphate-buffered saline (PBS); 7mM-Na2HPO4, 1·5mM-KH2PO4 (pH 7·4), 137 mM-NaCl, 2·7 mM-KCl) was injected, both intravenously and intraperitoneally, and the mouse was sacrificed on the fourth day after the injection.

Cell fusion and cloning

Dispersed spleen cells from an immunized mouse were mixed with PAI myeloma cells in a 5:1 ratio and fused in 1 ml 50% polyethylene glycol 4000 in serum-free RPMI medium, incubated for 1 min at 37°C and gradually diluted to 25 ml with serum-free RPMI medium. The cells were pelleted, resuspended in RPMI with 15% foetal calf serum, containing hypoxanthine, aminopterine, thymidine (HAT), distributed into ten 24-well plates (Costar 3524), each well containing l·5ml, and incubated; the medium was changed twice a week.

Hybridoma cells secreting antibodies were cloned two or three times by limiting dilution plating in 96-well plates (Falcon 3072) in RPMI medium. Culture supernatants were analysed for the presence of antibodies using an enzyme-linked immuno-absorbant assay (ELISA) and subsequently by immunoblot analysis as described below.

Selected cultures secreting specific type antibodies were used for growing ascitic tumours m old female Balb/c mice and the rest were frozen in 95% foetal calf serum, 5% dimethylsulphoxide, and stored in liquid nitrogen.

Detection of antibodies in the culture supernatant

ELISA

An enzyme-linked immuno-absorbant assay was used to detect hybrid cells secreting IgG antibodies. Microtiter plates (Flow Lab) were coated with 50 μl of proteins of the 20 S globin mRNP (4 μg ml−1) in PBS for 4h at room temperature. They were then saturated with 50 μl 3% bovine serum albumin (BSA) in PBS overnight at room temperature, stored at 4°C or used immediately for the test.

The hybridoma supernatant to be examined was added (50 μl per well) and the plates were incubated overnight at 18°C, or 4h at room temperature. The wells were washed, and then incubated with 50 μl peroxidase-coupled rabbit antimouse IgG (diluted 1: 1000 in 10% rabbit normal serum in PBS) for 3 h at room temperature. The wells were washed again, and finally treated with 50 μl of a solution containing 0·l % of 30% hydrogen peroxide, 2mM-ABTS (2,2′-azinodi)-(3-ethyl-benzothiazolin-sulphonate) into ELISA buffer (0·l M-sodium acetate, 0·05 M-NaH2PO4). Colour development was allowed to proceed for 30–60 min.

SDS-polyacrylamide gel electrophoresis (SDS-PAGE) and immunoblotting

Proteins were analysed by electrophoresis in one-dimensional 13% SDS-polyacrylamide gels, according to Laemmli (1970) and two-dimensional (non-equilibrium electrofocusing/SDS-polyacrylamide gel electrophoresis) as described by O’Farrell et al. (1977). Molecular weight markers were phosphorylase b (92K) bovine serum albumin (66K), ovalbumin (45 K) carbonic anhydrase (31K), soybean trypsin inhibitor (21K.) and lactalbumin (14K) (K = 103Mr).

Proteins from SDS—polyacrylamide gels were electrophoretically transferred to 0·45 μm nitrocellulose paper according to the slightly modified procedure of Towbin et al. (1979). The transfer buffer contained 25 mM-Tris HC1, 192 mM. glycine, 20% methanol and 0·01% SDS. Transfers were carried out in a transfer chamber at 500 mA for 2 h. Following transfer, the papers were air-dried and washed for 5 min in PBS containing 0·2% Tween 20 and 0·5% (w/v) BSA, and the unbound sites on the papers were blocked by overnight incubation at 4°C with 3% BSA in PBS. Excess BSA was removed by rinsing briefly in PBS. The papers were then incubated for 4h at room temperature, or overnight at 4°C, with the hybπdoma supernatant diluted in PBS (1:1), washed twice with PBS for 30 min, incubated with peroxidase-coupled rabbit anti-mouse IgG (1: 1000 in PBS containing 10% rabbit serum) and developed with 4-chloro-1-naphthol.

For the identification of antigens by the different monoclonal antibodies by one-dimensional gel electrophoresis, proteins of the 20 S duck globin mRNP were separated in a preparative gel and transferred to nitrocellulose. The paper was cut into narrow strips; such strips were incubated in sealed bags with individual MAbs, or with the polyclonal immune serum against total prosomes (dilution 1:300), using mouse non-immune serum as a control. For detection of antigens after two-dimensional gel electrophoresis, a single transfer paper was incubated several times with different MAbs consecutively.

Purification of immunoglobulins

Immunoglobulins were purified from ascitic fluid by precipitation with 50% saturated ammonium sulphate, followed by DEAE-Trisacryl chromatography, using 25 mM-Tris HC1, 35mM-NaCl (pH8·8) as elution buffer. The purity of the immunoglobulin was analysed by means of microzone electrophoresis on cellulose acetate strips.

Characterization of the monoclonal antibodies

The antibodies of the different subclones, amplified in ascitic fluid of mice, were purified by ammonium sulphate precipitation and DEAE-Trisacryl ion-exchange chromatography. The immunoglobulin subclass of the antibodies was determined by immunoblotting of hybridoma culture supernatants or purified antibodies, reacted with rabbit anti-mouse immunoglobulin subclasses (Gl, G2a, G2b, G3 and M). All secondary antibodies were from Miles. The results are summarized in Table 1. The anti-p27K antibody (IB5) and anti-p31K (AA4) were found to correspond to the IgGl subclass. The anti-p28K antibody (AB2) is of the IgG2a subclass, and the anti-p29K antibody (GD6) belongs to the IgG2b subclass.

Table 1.

Characterization of monoclonal antibodies prepared against prosome proteins

Characterization of monoclonal antibodies prepared against prosome proteins
Characterization of monoclonal antibodies prepared against prosome proteins

To determine the pl of each monoclonal antibody, isoelectric focusing in polyacrylamide gels was performed with purified antibodies. The four monoclonal antibodies focused as four principal bands. The isoelectric points were 5·9 for the anti-p27K (IB5), 7·0 for the anti-p28K (AB2), 6·5 for the anti-29K (GD6) and 5·2 for the p31K (AA4) component.

Isoelectric focusing (IEF)

Analytical isoelectric focusing of the purified immunoglobulin fraction from the ascitic fluid or [35S]methionine-labelled hybridoma supernatant was carried out as described by Pal & Modak (1984) on a LKB ‘Mutiphor’ apparatus using ampholines of pH 3·5–9·5. Electrofocusing was performed at 15 W (4h) at room temperature using 1 M-H3PO4 and 1 M-NaOH as anode and cathode electrolytes, respectively. The gel was fixed in 10% trichloroacetic acid for 1 h, and stained with 0·1% Coomassie Blue R-250 (in ethanol:acetic acid:water, 10:40:50, by vol.) for 30 min. For determining the pH gradient, either proteins of known pl were used, or the pH of the gel at different distances was measured as follows. A 1-cm wide strip was cut from one end of the gel before fixation and sliced into 1-cm pieces. The gel pieces were eluted in distilled water for several hours and the pH of the eluants was determined. A calibration curve for the pH gradient was plotted and the pl values of individual protein bands were determined by interpolation.

Indirect immunofluorescence microscopy of adult duck and embryonic chicken erythroblasts

Blood of anaemic ducks and embryonic chicken erythroblasts was smeared onto slides and fixed for 3 min in methanol: acetone (3:7, v/v) at −20°C. The slides were air dried for 15 min, quickly rinsed in PBS and refixed for 10 min in cold 2% paraformaldehyde in PBS containing 2mM-EDTA (or 10% normal duck serum, in the case of the adult duck erythroblasts). The cells were permeabilized with 0·l% Triton X-100 in PBS containing EDTA or 10% duck serum (3min), followed by a pre-treatment for 20min with 1% rabbit serum and 1% BSA, at room temperature. The slides were rinsed briefly in PBS and incubated with the different antibodies (ascitic fluid) diluted in PBS (1:100) for 2–3 h at room temperature in a humid chamber. Finally, the cells were incubated with the fluorescein isothiocyanate (FITC)-conjugated second antibody (rabbit anti-mouse IgG; 1:100). After each fixation and incubation with antibody, the slides were washed three times in PBS for 20 min each. The slides were mounted in 50% glycerol in PBS (pH8–5). Cells were examined with a Zeiss fluorescence microscope and photographed on Ilford HP5 film (400 ASA). Controls were carried out by the identical procedure, except that the antibody was replaced by mouse non-immune serum (diluted 1:50) or FITC-conjugated second antibody (1:50).

Electron microscopy and immunoperoxidase reaction

Erythroblasts of anaemic ducks smeared in plastic culture dishes were fixed and permeabilized as described for immunofluorescence microscopy. Thereafter, the endogenous peroxidase activity was blocked with 0·6% hydrogen peroxide in methanol for 30 min, followed by three washings of 10 min each. Pretreatment was with 1% rabbit normal serum and 1% BSA in PBS for 30 min at room temperature. The cells were incubated with the MAbs (ascitic fluid 1:50), polyclonal anti-prosome serum (1:100) and normal mouse serum (1:50), diluted in PBS containing 0·1% BSA for 2–3 h at room temperature. Subsequently, the peroxidase-coupled second antibody rabbit anti-mouse IgG (1:20) diluted in PBS containing 0·1% BSA was added for 30 min at room temperature. Cells were further washed in PBS and fixed with 2% glutaraldehyde in PBS for 15 min at room temperature, washed thoroughly, and incubated with 3,3′-diaminobenzidine tetrahydrochloride (0·5 mgml−1), and hydrogen peroxide in PBS for 30 min in the dark. After washing, the cells were prefixed in 2% osmium tetroxide for 30 min at room temperature, further dehydrated in ethanol and embedded in Epon while still attached to the plastic culture dish.

In the case of avian erythroblatosis virus (AEV)-transformed cells, the procedures for immunofluorescence and immuno-electron microscopy were the same, except that the fixation was done in 2% paraformaldehyde in PBS for 30 min, followed by permeabilization with 0·1% Triton X-100 for 15 min.

Immunogold procedure

AEV-transformed cells grown on plastic dishes were fixed in 2% paraformaldehyde in PBS for 30min. After washing in PBS (three washes of 10 min each), the cells were scraped, pelleted and embedded in Lowicryl K4M according to Carlemalm et al. (1980). Thin sections mounted on nickel grids were incubated for 1 h at room temperature with anti-28K (AB2) diluted 1:20 in PBS. Then, the grids were washed in PBS (three washes of 10 min each) and the cells were incubated with goat anti-mouse IgG conjugated to colloidal gold particles (I5nni) diluted 1:5 in PBS (Janssen Pharmaceutics). After successive washing in PBS, the grids were air-dried and stained with uranyl acetate.

Preparation of post-mitochondrial supernatant and total nuclear proteins of duck erythroblasts

Post-mitochondrial supernatant and nuclei of young erythroblasts from duck blood were prepared as described by Gander et al. (1973) and Maundrell & Scherrer (1979) respectively. Purified nuclei were extracted in buffer containing 0·14M-NaCl, 8mM-MnCIz, 10mM-EDTA, 40 μg ml−1 DNasel, 40 μgml−1 RNaseA, 1 mM-phenylmethylsulphonyl fluoride (PMSF) (1:1, v/v) by incubation for lh at 37°C. Samples of post-rnitochondrial supernatant and nuclear proteins were analysed on 13% polyacrylamide gels in SDS-containing buffer.

Molecular weight determination of the prosomal antigens recognized by monoclonal antibodies (MAbs)

The purified duck 19 S prosomes (Fig. 1A–C) isolated from free globin mRNPs were used as antigen to produce antibodies against prosomal proteins. This particle proved to be a good immunogen in mice: an adequate immune response and a high antibody titre were obtained after the third or fourth injection. After screening several mice, a single animal producing specific antibodies was chosen for the preparation of monoclonal antibodies. Of the several stable hybridoma cells, we chose four cell clones secreting different antibodies recognizing distinct duck prosomal proteins (Fig. 1). Antibodies secreted by four stable cellular sub-clones, named AA4, GD6, AB2 and IB5, recognize proteins of 31000, 29 000, 28 000 and 27000Mr, respectively. Note that in Fig. 1 all proteins identified by the selected MAbs are proteins of low molecular weight, characteristic of prosomal proteins, with no cross-reactivity to the other mRNP proteins (cf. Vincent et al. 1981). However, one antibody (GD6) detects more than one protein band in some mRNP preparations. As seen in Fig. 1H, it detects additional polypeptides of Mr 28K and 33K, although the stronger reaction is to the p29K component.

Fig. 1.

Detection of prosomal antigens by different monoclonal antibodies. A–C. Typical electron micrographs of prosome preparations from the duck globin mRNP. Bar, 20 nm. D–K. Proteins of the cytoplasmic duck globin mRNP (20 S complexes) were separated on SDS-polyacrylamide gels (13%) and electrophoretically transferred to nitrocellulose paper. D. Proteins of the partially purified globin mRNP (20S) complex stained with Coomassie Blue. E–J. Immunoblots of proteins seen in D and reacted with E, non-immune serum; F, mouse polyclonal anti-prosome serum; G–J, the different MAbs reacting with the prosomal antigens p31K (AA4), p29K (GD6), p28K (AB2) and p27K (1B5), respectively. K. Coomassie Blue staining of the proteins of the purified duck prosomes isolated from the globin mRNP and electrophoresed on a separate gel.

Fig. 1.

Detection of prosomal antigens by different monoclonal antibodies. A–C. Typical electron micrographs of prosome preparations from the duck globin mRNP. Bar, 20 nm. D–K. Proteins of the cytoplasmic duck globin mRNP (20 S complexes) were separated on SDS-polyacrylamide gels (13%) and electrophoretically transferred to nitrocellulose paper. D. Proteins of the partially purified globin mRNP (20S) complex stained with Coomassie Blue. E–J. Immunoblots of proteins seen in D and reacted with E, non-immune serum; F, mouse polyclonal anti-prosome serum; G–J, the different MAbs reacting with the prosomal antigens p31K (AA4), p29K (GD6), p28K (AB2) and p27K (1B5), respectively. K. Coomassie Blue staining of the proteins of the purified duck prosomes isolated from the globin mRNP and electrophoresed on a separate gel.

The presence and distribution of prosomal antigens in repressed mRNP

To test if these MAbs recognize specific proteins of prosomes, the purified 20 S globin mRNPs were subjected to centrifugation in 0·5 M-KC1 buffer on sucrose gradients. In this condition, the 20S globin mRNP splits into four subcomponents (Civelli et al. 1980; Vincent et al. 1980) with the dissociated prosomes sedimenting in the 19 S zone of the gradient (Fig. 2A). SDS-PAGE analysis of proteins from each fraction in this region of the gradient, shows a group of small proteins of Mr ranging between 21 000 and 35 000, and a 56000 Mr component characteristic of the avian prosome (Fig. 2B). As shown previously, the still higher Mr components are not part of the purified prosomes (Martins de Sa et al. 1986).

Fig. 2.

Dissociation of the globin mRNP and localization of the prosomal antigens. A. The 20 S globin mRNP (2·0 A260 units) from duck erythroblasts were dissociated by 0·5M-KCI and separated on isokinetic sucrose gradients (5% to 25%, w/w) in buffer containing 0·5 M-KC1 (Beckman rotor SW4l, 37 000 revs min−1, 9172h, 4°C), as described by Civelli et al. (1980). B. Samples of 15 μl from alternate fractions were pooled and precipitated with 10% trichloracetic acid (TCA) for 30 min on ice, washed with acetone, dried, electrophoresed on 13% SDS-polyacrylamide gels and stained with Coomassie Blue. C. Duplicate gels were electrotransferred onto nitrocellulose sheets and immunoreacted, successively with the four MAbs. Cl, anti-p27K (IB5); C2, anti-p31K (AA4); C3, anti-p28K (AB2); C4, anti-p29K (GD6).

Fig. 2.

Dissociation of the globin mRNP and localization of the prosomal antigens. A. The 20 S globin mRNP (2·0 A260 units) from duck erythroblasts were dissociated by 0·5M-KCI and separated on isokinetic sucrose gradients (5% to 25%, w/w) in buffer containing 0·5 M-KC1 (Beckman rotor SW4l, 37 000 revs min−1, 9172h, 4°C), as described by Civelli et al. (1980). B. Samples of 15 μl from alternate fractions were pooled and precipitated with 10% trichloracetic acid (TCA) for 30 min on ice, washed with acetone, dried, electrophoresed on 13% SDS-polyacrylamide gels and stained with Coomassie Blue. C. Duplicate gels were electrotransferred onto nitrocellulose sheets and immunoreacted, successively with the four MAbs. Cl, anti-p27K (IB5); C2, anti-p31K (AA4); C3, anti-p28K (AB2); C4, anti-p29K (GD6).

Electroblotting of proteins from duplicate gels onto nitrocellulose paper, and subsequent reaction with the different MAbs (Fig. 2C), confirmed the presence of prosomal proteins only in the 19 S region; no reaction was found elsewhere in the gradient. This confirms our previous results (Schmid et al. 1984; Martins de Sa el al. 1986) showing that the prosomes sediment as a compact particle of defined protein composition with a 19 S peak component.

Fig. 3 shows a two-dimensional gel electrophoretic analysis of polypeptides of prosomes isolated from purified duck globin mRNPs (Fig. 3A), and the identification of the antigens recognized by the four MAbs (Fig. 3B). Heterogeneity of all recognized antigens was observed when the proteins were separated by two-dimensional gel electrophoresis. Each ‘single’ polypeptide of defined Mr observed in onedimensional SDS-PAGE is resolved into several components with different pl. In the case of the 29K polypeptide recognized by MAb GD6, heterogeneity in both the pl and Mr can be observed. The same results were obtained when prosomes were directly dissolved in SDS prior to electrophoresis, or when the proteins were first separated from the RNA by extraction with hot phenol.

Fig. 3.

Identification of the prosomal antigens by immunoreaction after two-dimensional gel electrophoresis. A. Proteins of purified prosomes from globin mRNP of duck erythroblasts were separated by two-dimensional gel electrophoresis and stained with Coomassie Blue. B. The separated proteins were transferred onto nitrocellulose sheets and successively immunoreacted with the different monoclonal antibodies: Bl, anti-p31K (AA4); B2, anti-p27K (1B5); B3, anti-p28K (AB2); B4, anti-p29K (GD6). The antigens in the gels are indicated by the following symbols: (◂) 31K; (◃) 27K; (⇽) 28K; (←) 29K; (←) 33K (Mr×10−3).

Fig. 3.

Identification of the prosomal antigens by immunoreaction after two-dimensional gel electrophoresis. A. Proteins of purified prosomes from globin mRNP of duck erythroblasts were separated by two-dimensional gel electrophoresis and stained with Coomassie Blue. B. The separated proteins were transferred onto nitrocellulose sheets and successively immunoreacted with the different monoclonal antibodies: Bl, anti-p31K (AA4); B2, anti-p27K (1B5); B3, anti-p28K (AB2); B4, anti-p29K (GD6). The antigens in the gels are indicated by the following symbols: (◂) 31K; (◃) 27K; (⇽) 28K; (←) 29K; (←) 33K (Mr×10−3).

Comparison of the antigenic relatedness of the prosomal proteins of different species

Two-dimensional gel electrophoretic analysis of proteins from prosomes of duck, mouse and human cells led to the observation of an almost identical protein pattern in these three species (Martins de Sa et al. 1986). Here we extend this observation using the MAbs to detect the cross-reactivity of anti-duck prosome antibodies with antigens of several vertebrate species. The antigenic similarity between proteins of prosomes isolated from globin mRNPs of duck and mouse erythroblasts, from mRNPs of AEV-transformed chicken erythroblasts and HeLa cells, was analysed by immunoblotting of one-dimensional electrophoretograms (Fig. 4). The results show that the two antigens p27K and p31K (antibodies 1B5, AA4), have common epitopes in all four species analysed (Fig. 4B,E). The antigen p28K (MAb AB2) is detected only in avian species (Fig. 4C), although a similar protein of the same Mr and pl is detected by Coomassie Blue staining of proteins from the mouse and human species (Martins de Sa et al. 1986). The anti-p29K antibody designated GD6 (Fig. 4D) showed strong reaction with the corresponding antigen in the avian species, and a very faint one in HeLa cells; but it does not react with any polypeptide of the mouse prosomes.

Fig. 4.

Immunological cross-reactivity among prosome proteins obtained from different species. A. Coomassie Blue staining of the mRNP or prosomes of different species (duck, mouse, HeLa and chicken). B–E. Immunoblots of the same gel as A reacted with MAbs, anti-p27K (B), anti-p28K (C), anti-p29K (D), anti-p31K (E).

Fig. 4.

Immunological cross-reactivity among prosome proteins obtained from different species. A. Coomassie Blue staining of the mRNP or prosomes of different species (duck, mouse, HeLa and chicken). B–E. Immunoblots of the same gel as A reacted with MAbs, anti-p27K (B), anti-p28K (C), anti-p29K (D), anti-p31K (E).

Cellular distribution of prosome antigens in differentiating avian erythroblasts

Purified immunoglobulins or ascitic fluid containing MAbs, as well as polyclonal antibodies, were used to study the intracellular distribution of prosome proteins by indirect immunofluorescence microscopy. As a control the serum from a non-immunized mouse was used. The four different monoclonal antibodies directed against the different prosomal proteins showed a similar distribution in terminally differentiating duck erythroblasts. In all cases a reaction was found in the nuclei as well as the cytoplasm (Fig. 5). Interestingly, in the nuclei, some antigens seem more uniformly distributed and others are restricted to characteristic patches.

Fig. 5.

Immunolocalization of prosome antigens in avian erythroblasts by immunofluorescence with monoclonal antibodies. A-F. Duck erythroblasts in terminal differentiation. C, anti-p27K; D, anti-p28K; E, anti-p29K; F, anti-p3lK; B, polyclonal anti-prosome serum; A, control with mouse non-immune serum. G-I. Embryonic chicken erythroblasts reacted with anti-29K antibody. Cells from 3-day-old (G), 7-day-old (H), and 19-day-old (I) embryonic chicken. J. Negative control with mouse non-immune serum. ×2250.

Fig. 5.

Immunolocalization of prosome antigens in avian erythroblasts by immunofluorescence with monoclonal antibodies. A-F. Duck erythroblasts in terminal differentiation. C, anti-p27K; D, anti-p28K; E, anti-p29K; F, anti-p3lK; B, polyclonal anti-prosome serum; A, control with mouse non-immune serum. G-I. Embryonic chicken erythroblasts reacted with anti-29K antibody. Cells from 3-day-old (G), 7-day-old (H), and 19-day-old (I) embryonic chicken. J. Negative control with mouse non-immune serum. ×2250.

The presence and cytolocalization of prosomes seem to vary with the degree of maturation of erythroblasts into erythrocytes. The fully differentiated erythrocytes show very weak fluorescence, the residual signal being localized around the nuclear membrane. When the polyclonal antibody directed against total prosomes was used, the pattern of fluorescence was comparable to that obtained with the monoclonal antibodies (Fig. 5B). A very similar pattern was obtained with embryonic erythroblasts, which mature in a more or less synchronous pattern (Fig. 5G–J). Fig. 6 shows chicken erythroblasts transformed by the retrovirus AEV. In these cells also, prosomal antigens are detected in both cellular compartments, the nucleus and cytoplasm. In the former, the four different MAbs seem to show slightly different immunofluorescence pattern indicating a possible variation in the nuclear distribution of each antigen.

Fig. 6.

Immunofluorescence on AEV-transformed erythroblasts. A. Negative control using the second labelled antibody alone. B. Labelling for anti-p27K; C, labelling for anti-p28K; D, labelling for anti-p29K; E, labelling for anti-p31K. ×2000.

Fig. 6.

Immunofluorescence on AEV-transformed erythroblasts. A. Negative control using the second labelled antibody alone. B. Labelling for anti-p27K; C, labelling for anti-p28K; D, labelling for anti-p29K; E, labelling for anti-p31K. ×2000.

These immunofluorescence results were confirmed by immunoperoxidase staining and electron microscopy of normal and AEV-transformed avian erythroblasts. As clearly shown by comparison with the control (no MAb; Fig. 7A), the normal erythroblasts exposed to the various MAbs showed strong staining in the cytoplasm, and patches of staining in the nucleus (Fig. 7C–F). In the nucleus, the reaction is mostly concentrated in the nucleoplasm (large arrows), with lower intensity over the chromatin (small arrows) and the periphery of the chromatin was more intense when the polyclonal total anti-prosome serum was used (Fig. 7B). In addition, all four antibodies tested label the plasma membrane to various degrees. The nuclear membrane of these cells seems to be largely devoid of antigen, and regularly an empty gap is observed characteristic of normal erythroblasts. In contrast, in AEV-transformed cells (Fig. 7G–L), the nuclear membrane was intensely stained by all four MAbs tested; the prosome antigens were clearly contained in both cellular compartments, as in the normal erythroblasts (Fig. 7G–L).

Fig. 7.

1mmunocytochemical localization of prosomal antigen in normal and AEV-transformed avian erythroblasts. Cells were incubated with non-immune serum, polyclonal anti-prosome serum and the four anti-prosome MAbs, and processed for immunoperoxidase staining as described in Materials and methods. A–F. Erythroblasts from normal birds. A. Control, non-immune mouse serum; B, polyclonal anti-prosome antiserum. C–F. Monoclonal antibodies: C, anti-p27K; D, anti p28K; E, anti-p29K; F, anti-p31K. ch, chromatin; is, interchromatin space; nu, nucleoli; nl, nuclear lamina. × 16000. G–L. AEV-transformed erythroblasts. G, Non-immune serum; H, polyclonal antiserum. I–L. Monoclonal antibodies: 1, anti-p27K; J, anti-28K; K, anti-p29K; L, anti-p31K. × 18000.

Fig. 7.

1mmunocytochemical localization of prosomal antigen in normal and AEV-transformed avian erythroblasts. Cells were incubated with non-immune serum, polyclonal anti-prosome serum and the four anti-prosome MAbs, and processed for immunoperoxidase staining as described in Materials and methods. A–F. Erythroblasts from normal birds. A. Control, non-immune mouse serum; B, polyclonal anti-prosome antiserum. C–F. Monoclonal antibodies: C, anti-p27K; D, anti p28K; E, anti-p29K; F, anti-p31K. ch, chromatin; is, interchromatin space; nu, nucleoli; nl, nuclear lamina. × 16000. G–L. AEV-transformed erythroblasts. G, Non-immune serum; H, polyclonal antiserum. I–L. Monoclonal antibodies: 1, anti-p27K; J, anti-28K; K, anti-p29K; L, anti-p31K. × 18000.

Indirect immunoassay with colloidal gold-labelled anti-mouse antibodies carried out on sections of AEV-transformed erythroblasts further confirmed the localization of the prosome antigens. Gold-labelled antibodies detected the p28K antigen in both the nucleus and cytoplasm of these cells (Fig. 8). In the cytoplasm prosomes are everywhere, including the nuclear and the plasma membrane. In the nucleus, the higher resolution of this technique allows prosomal antigens to be clearly seen at the edge of the dense chromatin, where perichromatin granules are located (Puvion & Moyne, 1981). More extensive studies will be necessary to obtain further details and statistically valid data on the exact location of prosome antigens relative to chromatin.

Fig. 8.

Immunogold labelling of prosomal antigens (arrowheads) in sectioned chicken erythroleukaemic cells transformed by AEV, observed by electron microscopy. B. Cells were first labelled with the anti-p28K monoclonal antibody and then exposed to goat anti-mouse IgG conjugated to colloidal gold particles (15 nm diameter). A. Control with mouse non-immune serum, nu, nucleus. ×35000.

Fig. 8.

Immunogold labelling of prosomal antigens (arrowheads) in sectioned chicken erythroleukaemic cells transformed by AEV, observed by electron microscopy. B. Cells were first labelled with the anti-p28K monoclonal antibody and then exposed to goat anti-mouse IgG conjugated to colloidal gold particles (15 nm diameter). A. Control with mouse non-immune serum, nu, nucleus. ×35000.

Immunoblot analysis of the total protein extracts from duck erythroblast nuclei confirmed the presence of prosomal antigens in the nuclei (Fig. 9A). Immunoreaction with all four monoclonal antibodies shows specific antigen bands, of the same molecular weight as those observed in the cytoplasm.

Fig. 9.

Immunological detection of prosomal antigens in the nuclear sap and post-mitochondrial supernatant of duck erythroblasts. A. Nuclei were prepared and extracted as described in Materials and methods. The nuclease-treated proteins were separated by 13% SDS–PAGE, electroblotted onto nitrocellulose paper and reacted with monoclonal antibodies. Lane a, Coomassie Blue staining; lanes b—e, the same gel after electrotransfer onto nitrocellulose paper and immunoreaction with the antiprosome MAbs; lane b, anti-p27K; c, anti-p28K; lane d, anti-p29K; lane e, anti-p31K. B. Approximately 40 μg of post-mitochondrial supernatant was separated by 13% SDS-PAGE (containing twice the normal Tris HCl concentration to improve resolution), electroblotted onto nitrocellulose paper and reacted with a mixture of the MAbs (anti-p27K and anti-p3lK). Lane a, Coomassie Blue staining; lane b, immunoreaction with the anti-p27K and anti-p31K MAbs.

Fig. 9.

Immunological detection of prosomal antigens in the nuclear sap and post-mitochondrial supernatant of duck erythroblasts. A. Nuclei were prepared and extracted as described in Materials and methods. The nuclease-treated proteins were separated by 13% SDS–PAGE, electroblotted onto nitrocellulose paper and reacted with monoclonal antibodies. Lane a, Coomassie Blue staining; lanes b—e, the same gel after electrotransfer onto nitrocellulose paper and immunoreaction with the antiprosome MAbs; lane b, anti-p27K; c, anti-p28K; lane d, anti-p29K; lane e, anti-p31K. B. Approximately 40 μg of post-mitochondrial supernatant was separated by 13% SDS-PAGE (containing twice the normal Tris HCl concentration to improve resolution), electroblotted onto nitrocellulose paper and reacted with a mixture of the MAbs (anti-p27K and anti-p3lK). Lane a, Coomassie Blue staining; lane b, immunoreaction with the anti-p27K and anti-p31K MAbs.

To test if the prosome antibodies might cross-react with non-prosomal proteins in the cytoplasm, we fractionated proteins of the post-mitochondrial supernatant electrophoretically, blotted them onto nitrocellulose and exposed them to the antibodies. Fig. 9B shows that the anti-p27K and anti-p31K MAbs reacted only with the prosomal proteins of molecular weight 27 000 and 31000, respectively.

In previous publications, we have characterized the prosomes as a novel type of ubiquitous RNP particles, which were first observed as subcomplexes of free cytoplasmic mRNPs (Spohr et al. 1970; Vincent et al. 1980, 1983; Schmid et al. 1984; Martins de Sa et al. 1986). However, very little is known about the possible function of this class of ubiquitous highly conserved RNP particles.

To study their biological role in the cell, we produced monoclonal antibodies against duck prosomal proteins, and their cytolocalization was investigated. In particular, we address here the question of the cytolocalization of the prosome antigen during avian red cell differentiation. Of the four monoclonal antibodies (MAbs) produced, one MAb (anti-p28K) reacts only with the avian prosomes (duck and chick) whereas the others (anti-p27K, p29K and p31K) also recognize components of mouse and human prosomes. Recent results from our laboratory show that the p27K epitope is present in numerous other eukaryotic species including plants (M.-F. Grossi de Sa, unpublished results; Kremp et al. 1986). This suggests a basic function for the p27K protein that has been conserved in the course of evolution.

The heterogeneity in pI values observed for prosomal proteins, suggests that their function may be modulated by chemical modification. Contrary to antigens p27K, p31K and p28K, the p29K prosomal antigen also shows some heterogeneity of Mr. The presence of common epitopes on different polypeptides points to a recent evolutionary divergence of the polypeptides composing the prosome family, and the possible chemical diversity points to a complex function of the proteins in question.

Prosomes were initially separated from the ‘core mRNP’ by sedimentation in 0·5 M-KC1 sucrose gradients, and we used our MAbs to verify whether the prosomal proteins were only present at the level of the 19 S particle, or if they were distributed quite uniformly on a sucrose gradient. The four MAbs tested gave a clear answer: prosomal antigens were observed exclusively in the 19 S prosome peak and were not found in other fractions of the sucrose gradients. This result confirms that prosomes are likely to be individual stable particles, associated in a reversible fashion with the core mRNP.

The cytological data presented here indicate that the prosomal antigens also exist in the nucleus, and that they are probably at the nuclear membrane of AEV-transformed erythroleukaemic cells. The peroxidase label as well as immunofluorescence, shows the presence of prosome antigens in the nucleus, and the immunogold assays on sections confirm this observation. Particularly interesting is the presence of immunogold particles at the nuclear membrane (Fig. 8), a situation possibly related to the classical observations of RNPs penetrating the nuclear membrane (Stevens & Swift, 1966).

Further studies will be necessary to explore in more detail the present observation of prosome antigens in relation to cytologically known landmarks. Detailed studies on the nature of the prosome antigens and their interaction with chromosomes, nuclear matrix and the cytoskeleton are in progress (M. F. Grossi de Sa, unpublished observations). The biochemical and cytological data reported here make it likely that the prosome antigens may be everywhere in the cell and might interact with both mRNA (Martins de Sa et al. 1986) and its nuclear precursor forms (unpublished observations).

The data shown here give more substance to the possibility that the prosomes are related to other particles of similar morphology described in the literature (see Martins de Sa et al. 1986). In particular, the observation by Kleinschmidt et al. (1983) of ‘cylindershaped’ particles in the nuclei of Xenopus oocytes, and in the ooplasm can be related to ours, although those particles seem to be devoid of RNA, and their proteins are mainly located in the nuclei (Hugle el al. 1983). Particles of similar morphology have recently been isolated from the nuclei of Xenopus ovary cells by Castano et al. (1986) and shown to co-purify with a pre-tRNA processing activity. Although it is difficult at present to decide conclusively on the basis of these data if the morphological entity resembling prosomes, and the enzymic activity identified, are identical and exclusive, this observation could indicate that prosome-like particles might assume different functions in the nucleus and cytoplasm. This is known for instance in the case of (pre-) ribosomal and (pre-) messenger RNA and their associated proteins. Further investigation may reveal whether the nuclear particles observed by others are identical to prosomes, or whether they represent just structurally related RNPs.

The possibility existed, therefore, that the prosome distribution might also reflect the population of specific types of mRNA in a given cell. Indeed each type of differentiated cell contains specific sets of mRNAs, and recent data indicate that at least some of these mRNAs are not evenly distributed, but accumulate in specific sectors of the cell (see e.g. Lawrence & Singer, 1986; Colman et al. 1982). Of particular interest to us is the question of whether nuclear prosomes might exist in association with pre-mRNA.

One final point to be discussed here concerns the distribution of prosomal antigens in relation to differentiation. Evidence was obtained that the presence and distribution of prosomes varies with the stage of differentiation of erythropoietic cells. Mature erythrocytes seem no longer to possess prosome antigens, the last location being at the nuclear lamina. In the precursor cells, the stronger fluorescence indicates a predominantly cytoplasmic location for the prosomes, although the different antigens are also present in the nucleus in variable amounts, forming specific patterns, varying during the process of erythrocyte maturation. These observations suggest that immunophenotyping of cells according to their state of differentiation may become possible using the prosomal MAbs.

In conclusion, this investigation confirms the existence of prosomes in repressed mRNPs as a stable complex that can be dissociated from the free mRNPs, without even partial destruction. The monoclonal antibodies used allowed us to show a high but variable evolutionary conservation of the prosomal proteins. Of particular importance seems the observation of prosomal antigens in the nucleus and on the cellular membranes; this points to prosomal functions beyond the mere frame of the repressed mRNP. Finally, the differential distribution and loss of individual prosomal antigens during differentiation points to a possible implication of prosomes in differential gene control (Scherrer, 1980), along with the messenger and possibly pre-messenger ribonucleoproteins to which they associate. A wide field of further investigations may also open up with the possibility of using prosomal antibodies as markers of differentiation.

We are grateful to Dr J.-F. Bun for discussion and to Dr H.-P. Schmid for a kind gift of mouse prosomes, to Dr A. Vincent for critical reading, and to C. Cuisinier and R. Schwartzmann for help in the preparation of the manuscript. We thank M. Huesca and H. Grimal for excellent technical assistance. This research was supported by the French CNRS, the INSERM, the Ministère de la Recherche, the Association pour la Recherche sur le Cancer, and the Fondation pour la Recherche Médicale. C. Martins de Sa was supported by a fellowship from the Brazilian Conselho Nacional de Desenvolvimento Cientifico e Tecnologico (CNPq) and M.-F. Grossi de Sa by a fellowship from the Association pour la Recherche sur le Cancer and Brazilian Conselho Nacional de Desenvolvimento Cientifico e Tecnologico (CNPq).

Akhλyat
,
O.
(
1986
).
Etude du rôle des particuies ribonucléoprotéiques dans la régulation post-transcriptionnelle de (‘expression des gènes chez les eucaryotes. Doctoral thesis, University Paris VII
.
Akhayλt
,
O.
,
Infante
,
A. A.
,
Infante
,
D.
,
Martins de Sa
,
C.
,
Grossi de Sa
,
M.-F.
&
Scherrer
,
K.
(
1987
).
A particular prosome composed of ScRNA and multimers of a 21K protein inhibits synthesis in vitro
.
Eur.J. Biochem (in press)
.
Akhayat
,
O.
,
Vincent
,
A.
,
Goldenberg
,
S.
,
Person
,
A.
&
Scherrer
,
K.
(
1983
).
The translation of the messenger for the poly(A) binding protein associated with translated mRNA is suppressed
.
FEBS Lett
.
162
,
25
31
.
Brawerman
,
G.
(
1974
).
Eukaryotic pre-messenger RNA. A
.
Rev. Biochem
.
43
,
621
642
.
Carlemalm
,
E.
,
Villiger
,
W.
&
Acetarin
,
J. D.
(
1980
).
Advances in specimen preparation for electron microscopy. Novel low temperature embedding resins and a reformulated vespotal
.
Experientia
36
,
740
742
.
Castano
,
J. G.
,
Ornberg
,
R.
,
Koster
,
J. G.
,
Tobian
,
J. A.
&
Zasloff
,
M.
(
1986
).
Eukaryotic pre-tRNA 5’ processing nuclease: copurification with a complex cylindrical particle
.
Cell
46
,
377
387
.
Civelli
,
O. C.
,
Vincent
,
A.
,
Buri
,
J.-F.
&
Scherrer
,
K.
(
1976
).
Evidence for a translational inhibitor linked to globin mRNA in untranslated free cytoplasmic messenger ribonucleoprotein complexes
.
FEBS Lett
.
72
,
71
76
.
Civelli
,
O. C.
,
Vincent
,
A.
,
Maundrell
,
K.
,
Buri
,
J.-F.
&
Scherrer
,
K.
(
1980
).
The translational repression of globin mRNA in free cytoplasmic ribonucleoprotein complexes
.
Eur. J. Biochem
.
107
,
577
585
.
Colman
,
D. R.
,
Kreibich
,
G.
,
Fey
,
A. B.
&
Sabatini
,
D. D.
(
1982
).
Synthesis and incorporation of myelin polypeptides into CNS myelin
.
J. Cell Biol
.
95
,
598
607
.
Gander
,
E. S.
,
Stewart
,
A.
,
Morel
,
C.
&
Scherrer
,
K.
(
1973
).
Isolation and characterization of ribosome free cytoplasmic messenger ribonucleoprotein complexes from avian erythroblasts
.
Eur. J. Biochem
.
38
,
443
452
.
Horsch
,
A.
,
Spindler
,
E.
,
Martins de Sa
,
C.
,
Koehler
,
K.
&
Schmid
,
H.-P.
(
1987
).
Prosome discriminate between mRNA of adenovirus infected and uninfected HeLa cells in vitro
.
Eur. J. Cell Biol
.
43
,
27
.
Hugle
,
B.
,
Kleinschmidt
,
J. A.
&
Franke
,
W. W.
(
1983
).
Cylinder particles of Xenopus laevis. II. Immunological characterization and localization of their proteins in tissues and cultured cells
.
Eur. J. Cell Biol
.
32
,
157
163
.
Imaizumi-Scherrer
,
M.-T.
,
Maundrell
,
K.
,
Civelli
,
O. C.
&
Scherrer
,
K.
(
1982
).
Transcriptional and post-transcriptional regulation in duck erythroblasts
.
Devl Biol
.
93
,
126
138
.
Kleinschmidt
,
J. A.
,
Hugle
,
B.
,
Grund
,
C.
&
Franke
,
W. W.
(
1983
).
Cylinder particles of Xenopus laevis. I. Biochemical and electron microscopy characterization
.
Eur. J. Cell Biol
.
32
,
143
146
.
Kremp
,
A.
,
Schliephacke
,
M.
,
Kull
,
U.
&
Schmid
,
H.-P.
(
1986
).
Proxomes exist in plant cells too
.
Expl Cell Res
.
166
,
553
557
.
Laemmli
,
U. K.
(
1970
).
Cleavage of structural proteins during the assembly of the head of bacteriophage T4
.
Nature, Land
.
721
,
680
685
.
Lawrence
,
J. B.
&
Singer
,
R. H.
(
1986
).
Intracellular localization of messenger RNAs for cytoskeletal proteins
.
Cell
45
,
407
415
.
Martins de Sa
,
C.
,
Grossi de Sa
,
M.-F.
,
Akhayat
,
O.
,
Broders
,
F.
,
Horsch
,
A.
,
Schmid
,
H.-P.
&
Scherrer
,
K.
(
1986
).
Prosomes: ubiquity and inter-species structural variation
.
J. niolec. Biol
.
187
,
479
493
.
Maundrell
,
K.
,
Imaizum-Scherrer
,
M.-T.
,
Maxwell
,
S.
,
Civelli
,
O.
&
Scherrer
,
K.
(
1983
).
Messenger RNA for the 73,000 dalton poly(A)-binding protein occurs as translationally repressed mRNP in duck reticulocytes
.
J. biol. Chem
.
258
,
1387
1390
.
Maundrell
,
K.
&
Scherrer
,
K.
(
1979
).
Characterization of pre-messenger RNA-containing nuclear ribonucleoprotein particles from avian erythroblasts
.
Eur. J. Biochein
.
99
,
225
238
.
Morel
,
C.
,
Kayibanda
,
B.
&
Scherrer
,
K.
(
1971
).
Proteins associated with globin messenger RNA in avian erythroblasts: isolation and comparison with the proteins bound to nuclear messenger like RNA
.
EEBS Lett
.
18
,
54
88
.
O’farrell
,
P. Z.
,
Goodman
,
H.
&
O’farrell
,
P. H.
(
1977
).
High resolution two-dimensional electrophoresis of basic as well as acidic proteins
.
Cell
12
,
1133
1142
.
Pal
,
J. K.
&
Modak
,
S. P.
(
1984
).
Immunochemical characterization and quantitative distribution of crystallins in the epithelium and differentiating fibers cell populations of chick embryonic lens
.
Expl Eye Res
.
39
,
415
434
.
Puvion
,
E.
&
Moyne
,
G.
(
1981
).
In situ localisation of RNA structures
. In
The Cell Nucleus
, vol.
8
(ed.
H.
Busch
), pp.
59
115
.
New York
:
Academic Press
.
Scherrer
,
K.
(
1973
).
Messenger RNA in eukaryotic cells: The life-history of duck globin messenger RNA
. In
Protein Synthesis in Reproductive Tissues, Sixth Karolinska symposium on research methods in reproductive endocrinology
(ed.
E.
Diczfalusy
), pp.
95
-
129
. Copenhagen: Bogkrykkeritet Forum.
Scherrer
,
K.
(
1980
).
In Eukaryotic Gene Regulation
, vol.
1
(ed.
G. M.
Kolodny
), pp.
57
129
.
Boca Raton, Florida
:
CRC Press
.
Schmid
,
H.-P.
,
Akhayat
,
O.
,
Martins de Sa
,
C.
,
Puvion
,
F.
,
Koehler
,
K.
&
Scherrer
,
K.
(
1984
).
The prosome: an ubiquitous morphologically distinct RNP particules associated with repressed mRNPs and containing ScRNA and containing ScRNA and a characteristic set of proteins
.
EM BO J
.
3
,
29
34
.
Spirin
,
A. S.
(
1969
).
Informosomes
.
Eur. J. Biochein
.
10
,
20
35
.
Spirin
,
A. S.
,
Belγrisina
,
N. V.
&
Ajtkhozhin
,
M. A.
(
1964
).
Messenger RNA in early embryogenesis
.
J. gen. Biol
.
25
,
321
337
.
Spohr
,
G.
,
Granbouland
,
N.
,
Moul
,
C.
&
Scherrer
,
K.
(
1970
).
Messenger RNA in HeLa cells: An investigation of free and polyribosome bound cytoplasmic messenger RNP particles by kinetic labelling and electron microscopy
.
Eur. J. Biochein
.
17
,
296
318
.
Stevens
,
B. J.
&
Swift
,
H.
(
1966
).
RNA transport from nucleus to cytoplasm in Chimnoinits salivary glands
.
J. Cell Biol
.
31
,
55
77
.
Towbin
,
H.
,
Staehelin
,
T.
&
Gordon
,
J.
(
1979
).
Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications
.
Proc. natn. Acad. Sci. U.S.A
.
76
,
4350
4354
.
Vincent
,
A.
,
Akhayat
,
O.
,
Goldenberg
,
S.
&
Scherrer
,
K.
(
1983
).
Differential repression of specific mRNA in erythroblasts cytoplasm: a possible role for free mRNP particles
.
EMBO J
.
2
,
1869
1876
.
Vincent
,
A.
,
Civelli
,
O. C.
,
Maundrell
,
K.
&
Scherrer
,
K.
(
1980
).
Identification and characterisation of the translationally repressed cytoplasmic globin messenger ribonucleoprotem particles from duck erythroblasts
.
Eur. J. Biochein
.
112
,
617
633
.
Vincent
,
A.
,
Goldenberg
,
S.
,
Standart
,
N.
,
Civelli
,
O.
,
Imaizumi-Scherrer
,
M.-T.
,
Maundrell
,
K.
&
Scherrer
,
K.
(
1981
).
Potential role of mRNP proteins in cytoplasmic control of gene expression in duck erythroblasts
.
Molec. Biol. Rep
.
7
,
71
81
.
Vincent
,
A.
,
Civelli
,
O.
,
Buri
,
J.-F.
&
Scherrer
,
K.
(
1977
).
Correlation of specific coding sequences with specific proteins associated in untranslated cytoplasmic messenger ribonucleoprotein complexes of duck erythroblasts
.
EEBS Lett
.
77
,
281
286
.