THE PREPARATION AND ULTRASTRUCTURE OF AVIAN ERYTHROCYTE NUCLEAR ENVELOPE ENCLOSED BY THE PLASMA MEMBRANE

Transcription

1 J. Cell Sci. 34, (1978) 8l Printed in Great Britain Company of Biologists Limited igj8 THE PREPARATION AND ULTRASTRUCTURE OF AVIAN ERYTHROCYTE NUCLEAR ENVELOPE ENCLOSED BY THE PLASMA MEMBRANE JAMES R. HARRIS Biomembrane Unit, Division of Biochemistry, North East London Polytechnic, Romford Road, London E\ 5 4LZ, England SUMMARY A procedure is described for the preparation of avian erythrocyte nuclear envelope ghosts which remain enclosed by the ellipsoid plasma membrane. Haemoglobin-free nucleated chicken erythrocyte ghosts are treated in a low ionic strength buffer plus heparin which brings about decondensation of the chromatin. This is followed by solubilization of the chromatin by digestion with pancreatic deoxyribonuclease-1. When studied by light microscopy using either phase-contrast or Nomarski interference optics, the ellipsoid plasma membrane is clearly seen to remain with the collapsed nuclear envelope trapped inside. This interpretation is supported by negative-staining electron microscopy using ammonium molybdate, which in addition reveals the presence of the nuclear pore complexes. The suggestion is advanced that structural protection is provided for the fragile nuclear envelope system by the surrounding plasma membrane, which might account for the final nuclear envelope being in the form of relatively intact ghosts with well denned nuclear pore complexes. The nuclear envelope is highly fragmented when the plasma membrane is absent, the nuclear pore complexes showing appreciable breakdown. Thin sectioning supports the results of negative staining and in addition shows the nuclear envelope retained within the plasma membrane to be composed of both inner and outer nuclear membranes, but the nuclear pore complexes are not clearly defined. INTRODUCTION The procedure previously used for the production of haemoglobin-free avian erythrocyte ghosts (Harris & Brown, 1971a, b), which retain their nuclei has been further developed in an attempt to release the chromatin from the nuclei retained within the plasma membrane ghosts. It was hoped that the morphological juxtaposition of the 2 nuclear membranes and the integrity of the nuclear pore complexes might be retained under these circumstances. The direct treatment of isolated avian erythrocyte nuclei in an attempt to obtain intact nuclear envelopes (Harris, Agutter & Mime, 1978) has proved to be only partially successful, in that the final product consists primarily of fragmented nuclear envelope containing damaged nuclear pore complexes. The protective action afforded by the plasma membrane surrounding the fragile nuclear envelope throughout the low ionic strength swelling in the presence of heparin and the deoxyribonuclease digestion used to release the chromatin, is considered to be of prime importance. This is particularly so since the overall treatment is basically the same as that applied to isolated nuclei. It will be shown that the nuclear envelope

2 82 J. R. Harris enclosed by plasma membrane produced by this treatment is in a superior morphological condition to that isolated directly from nuclei (Harris et al. 1978). Despite the fact that the procedure presented below does not produce nuclear envelope as highly purified as that from isolated nuclei, it is clearly shown that it does provide material suitable for detailed electron-microscopic investigations of the extremely fragile nuclear envelope of the avian erythrocyte,which unfortunately cannot be quantitatively obtained in the form of purified intact nuclear envelope ghosts, as is possible from other tissues (Harris, 1978; Harris & Milne, 1974; Kay, Fraser & Johnston, 1972). Several investigators other than the author and his colleagues have performed biochemical studies on avian erythrocyte plasma membrane and nuclear envelope (Zentgraph, Deumling, Jarasch & Franke, 1971; Shelton, 1973; Blanchet, 1974; Jackson, 1975; Jackson, 1976a, b; Caldwell, 1976; Weise & Ingram, 1976; Shelton, Cobbs, Povlishock & Burkat, 1976). It is clear that in most cases the plasma membrane is separated from the nucleus by procedures such as homogenization, ultrasonication and nitrogen cavitation. Although Jackson (1975) did perform studies with deoxyribonuclease-treated nucleated chicken erythrocyte ghosts, he did not include any electron micrographs of his material. The available publications dealing with avian erythrocyte nuclear envelope which do include electron micrographs support the interpretation that this membrane system is very fragile. The inner and outer nuclear membrane and nuclear pore complexes are generally not definable by thin sectioning of isolated nuclear envelope fragments, but Harris et al. (1978) clearly show the location of damaged nuclear pore complexes by negative staining. EXPERIMENTAL METHODS Preparation of nuclear envelope within plasma membrane Chicken erythrocyte ghosts were prepared by the method of Harris & Brown (1971 a, b). The treatment of these nucleated haemoglobin-free erythrocyte ghosts with deoxyribonuclease permits the release of solubilized chromatin from the nuclei, which in turn escapes from the cytoplasmic compartment through lesions in the surrounding plasma membrane. Erythrocyte ghosts from 40 ml of packed cells were washed twice at 3000 rev/min (1000 j ) for 5 min with 10 mm Tris-HCl buffer (ph 70). The slightly gelatinous pellets were then pooled and made up to a volume of 400 ml with the Tris-HCl buffer. To disperse the chromatin further, heparin was added to a concentration of 20 U./ml and deoxyribonuclease-1 (Sigma DN-100) was added to give a concentration of 20 /tg per ml. The gel dispersed rapidly, and to activate the enzyme 0-04 ml of 10 M magnesium chloride was added, giving a final concentration of o-i truii. After 10 min incubation at room temperature with gentle stirring, the suspension was centrifugally pelleted and washed with 10 mm Tris-HCl (ph 7-4), twice at rev/min (12000 g) for Fig. 1. Chicken erythrocyte plasma membrane ghosts containing nuclear envelope ghosts, as observed by light microscopy using, A, Nomarski interference optics, compared with the initial nucleated erythrocyte ghosts, B. x Fig. 2. Chicken erythrocyte plasma membrane ghosts containing nuclear envelope ghosts, as observed at low electron-optical magnification. Negatively stained with 2 % ammonium molybdate (ph 7-0). Arrowheads indicate the location of the nuclear envelopes, x 8000.

3 Avian erythrocyte nuclear envelope

4 84 J. R. Harris 10 min and 6 times at rev/min (48000 g) for 5 min to remove most of the solubilized chromatin. The membrane material was then purified by isopycnic banding on a 10 to 20 M sucrose gradient prepared in 10 mm Tris-HCl (prf 74). Centrifugation was performed for 16 h at rev/min (83000 g). The erythrocyte plasma membrane ghosts containing nuclear envelope banded at a mean density of 1-18, were removed and washed centrifugally in 10 mm Tris-HCl buffer (ph 70) to remove the sucrose prior to performing processing for electron microscopy. Electron microscopy Negatively stained specimens were prepared on carbon-coated grids using the nuclear envelope material directly after the removal of sucrose by centrifugal washing. The single-drop technique (Harris & Agutter, 1970) was used for applying the membrane suspension and negative stain (2 % ammonium molybdate, ph 7-0) to the grids. For thin sectioning the membrane material was fixed overnight at room temperature in Dalton's chrome-osmium (Dalton, 1955), dehydrated with graded ethanols and epoxypropane before embedding in TAAB epoxy resin. Thin sections were cut using glass knives with a Reichert OM U2 ultramicrotome and were stained with uranyl acetate and lead citrate. Specimens were studied in a Philips EM 301S electron microscope at an accelerating voltage of 80 kv. Photographs were taken on Ilford 3 J x 4 in. (8-25 x 102 cm) plates, type EM 6. RESULTS AND DISCUSSION The erythrocyte plasma membrane ghosts containing nuclear envelope ghosts readily seen by light microscopy using phase contrast or Nomarski interference optics, see Fig. 1. Electron microscopy confirms and extends the light-microscopic observations. By negative staining, the ellipsoid outline of the plasma membrane is clearly revealed as is the position of the nuclear envelope trapped inside, see Fig. 2. Small, dense mitochondria can be seen adjacent to the nuclear envelope. If the nuclear envelope is adequately spread rather than folded, it is possible to see the distribution of the nuclear pore complexes over the whole area of the flattened envelope (Fig. 3). At higher electron-optical magnifications (Fig. 4) it can be seen that the nuclear pore complexes possess pronounced annuli with indication of the 8 annular subunits. It is proposed that the apparent morphological integrity of the overall nuclear envelope and the nuclear pore complexes under these conditions may well be due to the structural protection afforded by the surrounding plasma membrane throughout the isolation procedure. Nevertheless, a few free nuclear envelopes were always detected (Fig. 5), which may be derived from nuclei released during the haemolysis of the erythrocytes. It might be expected that the ultrastructural detail of the free nuclear envelope as revealed by negative staining would be significantly better than that of the nuclear envelope entrapped by the plasma membrane, owing to the absence of the plasma membrane layer above and below the nuclear envelope. On close comparison Fig. 3. Part of a chicken erythrocyte plasma membrane ghost containing a nuclear envelope ghost which is adequately flattened so that the distribution of the nuclear pore complexes can be clearly seen. Arrowheads indicate densely stained mitochondria. Negatively stained with 2 % ammonium molybdate. x Fig. 4. Part of the nuclear envelope ghost shown in Fig. 3, revealing detail of the nuclear pore complexes. Negatively stained with 2 % ammonium molybdate. x

5 ; * * Avian erythrocyte nuclear envelope I- t V

6 J. R. Harris

7 Avian erythrocyte nuclear envelope 87 of the images of the entrapped nuclear envelope (Figs. 3, 4) with that of the free nuclear envelope (Fig. 5), and with the nuclear envelope obtained directly from purified nuclei by the swelling treatment in the presence of heparin followed by deoxyribonuclease digestion (Harris et al. 1978) as shown in Fig. 6, it is apparent that the nuclear envelope in Figs. 5 and 6 is considerably more damaged than that in Figs. 3 and 4, but that the nuclear envelope surface shows only marginally greater detail. What is of greater significance is the distortion of the nuclear pore complexes and the apparent loss of much of the annular material in Figs. 5 and 6. It should be emphasized again that the preparative procedure applied to purified nuclei does produce nuclear envelope in a predominantly fragmented state, rather than as nuclear envelope ghosts. When thin-sectioned, the nuclear envelope entrapped within the plasma membrane can be seen to retain both the inner and outer nuclear membranes (Figs. 7, 8). The outer nuclear membrane is extremely tenuous and the inner nuclear membrane appears denser due to residual chromatin particles attached to its nucleoplasmic surface. This chromatin could be intimately associated with the inner nuclear membrane, but it is more likely to be a contaminant due to incomplete release during the deoxyribonuclease digestion. The location of the nuclear pore complexes is difficult to define by thin sectioning owing to the low pore density, as shown in Figs This reduces the chance of any one nuclear pore complex being located within the plane of the section under study, while at the same time being oriented appropriately to give the circular image of a tangentially sectioned pore complex or the 'press-stud' image of the perpendicularly sectioned pore complex (Figs. 7, 8). The use of heparin for the release of chromatin from nuclei was introduced by Bornens (1973) and has recently been extended as a detailed production of nuclear envelope from rat liver nuclei (Bornens & Courvalin, 1978). One point of significance that emerges from this work is that the nuclear pore complex annuli are completely removed, resulting in thin-sectioned rat liver nuclear envelope showing smooth-edged pores with no indication of the annular subunits. Hildebrand & Okinaka (1976), who applied a heparin treatment to cultured cell nuclei, commented that their fragmented nuclear envelope contained few nuclear pore complexes, whereas they were clearly distinct on the intact nuclei, again indicating the disruptive action of heparin. Despite this criticism of the use of heparin it appears from the results presented above, that when the fragile avian erythrocyte nuclear envelope is protected by the surrounding plasma membrane throughout the preparative procedures, the effect of heparin is less marked than is the case when isolated nuclei are treated by the same procedure. The availability of relatively intact avian erythrocyte nuclear envelope ghosts Fig. 5. An isolated nuclear envelope ghost, free from plasma membrane. The nuclear pore complexes appear to be distorted and have less annular material than those visible in Figs. 3, 4. Negatively stained with 2 % ammonium molybdate. x Fig. 6. A nuclear envelope ghost prepared from purified nuclei. Note the similarity to the envelope in Fig. 5, in particular the presence of distorted nuclear pore complexes deficient in annular material. Negatively stained with 2 % ammonium molybdate. x

8 88 J. R. Harris onm T /A7A77 v V \ inm f 8 yo/77 onm Figs. 7, 8. Thin-sectioned chicken erythrocyte plasma membrane ghosts containing nuclear envelope ghosts. Residual chromatin is seen to be adhering to the nucleoplasmic surface of the inner nuclear membrane (inm). The tenuous outer nuclear membrane (onm) is visible in places. The plasma membrane is indicated by pm. x and 78000, respectively.

9 Avian erythrocyte nuclear envelope 89 enclosed by plasma membrane ghosts may provide a system for investigating the permeability of the nuclear envelope to macromolecules. The system may similarly be of application to cell fusion studies and nuclear reactivation following fusion, where it is of interest to know whether the nuclear pore complex density increases on reactivation of the avian erythrocyte RNA metabolism. There is also considerable scope for further ultrastructural and biochemical study, particularly in relation to cellular development of the avian erythrocyte, membrane receptors and the nuclear envelope pore complex-lamina which has been shown by Aaronson & Blobel (1975) to remain after the extraction of rat liver nuclear envelope with the non-ionic surfactant Triton X-100 and high salt concentration. This latter aspect has been investigated electrophoretically by Shelton (1976). Although preservation of the morphological integrity of the nuclear envelope and the nuclear pore complexes has been emphasized throughout this publication, it must be stated that there is considerable value in having available highly purified yet fragmented nuclear envelope containing damaged pore complexes, for comparative biochemical studies, particularly if these fragments can be classified as being predominantly inner or outer nuclear membrane. This work has been, supported by grants from the Medical Research Council and The Royal Society. The technical assistance of Mr J. Murdock and Mr J. Kerr is gratefully acknowledged. REFERENCES AARONSON, R. P. & BLOBEL, G. (1975). Isolation of nuclear pore complexes in association with a lamina. Proc. natn. Acad. Sci. U.S.A. 72, BLANCHET, J. P. (1974). Chicken erythrocyte membranes: comparison of nuclear and plasma membranes from, adults and embryos. Expl Cell Res. 84, BORNBNS, M. (1973). Action of heparin on nuclei: solubilization of chromatin enabling the isolation of nuclear membranes. Nature, Lond. 244, BORNENS, M. & COURVALIN, J. C. (1978). Isolation of nuclear envelopes with polyanions. J. Cell Biol. 76, CALDWELL, A. B. (1976). Proteins of the turkey erythrocyte membrane. Biochemistry, N.Y. 15, DALTON, A. J. (1955). A chrome-osmium fixative for electron microscopy. Anat. Rec. 121, 281 (Abstr.). HARRIS, J. R. (1978). The biochemistry and ultrastructure of the nuclear envelope. Biochim. biophys. Ada 515, 55~i 4- HARRIS, J. R. & AGUTTER, P. S. (1970). A negative staining study of human erythrocyte ghosts and rat liver nuclear membranes..7. Ultrastruct. Res. 33, HARRIS, J. R., AGUTTER, P. S. & MILNE, J. F. (1978). Ultrastructural and biochemical studies on avian erythrocyte plasma membrane and nuclear envelope. Micron (in press). HARRIS, J. R. & BROWN, J. N. (1971a). The preparation of nucleated erythrocyte ghosts from avian erythrocytes. Br. Poult. Sci. 12, HARRIS, J. R. & BROWN, J. N. (1971 b). Fractionation of the avian erythrocyte: an ultrastructural study. J. Ultrastruct. Res. 33, HARRIS, J. R. & MILNE, J. F. (1974). A rapid procedure for the isolation and purification of rat liver nuclear envelope. Trans. Biochem. Soc HILDEBRAND, C. E. & OKINAKA, R. T. (1976). A rapid method for preparation of nuclear membranes from mammalian cells. Analyt. Biochem. 75, JACKSON, R. C. (1975). The exterior surface of the chicken erythrocyte. J. biol. Chevi. 250,

10 90 J. R. Harris JACKSON, R. C. (1976a). Polypeptides of the nuclear envelope. Biochemistry, N.Y. 15, JACKSON, R. C. (19765). On the identity of nuclear membrane and non-histone nuclear proteins. Biochemistry, N.Y. 15, KAY, R. R., FRASER, D. & JOHNSTON, I. R. (1972). A method for the rapid isolation of nuclear membranes from rat liver. Eur. J. Biochem. 30, SHELTON, K. R. (1973). Plasma membrane and nuclear proteins of the goose erythrocyte. Can. J. Biochem. 51, SHELTON, K. R. (1976). Selective effects of non-ionic detergent and salt solutions in dissolving nuclear envelope proteins. Biochbn. biophys. Acta 455, SHELTON, K. R., COBBS, C. S., POVLISHOCK, J. T. & BURKAT, R. K. (1976). Archs Biochem. Biophys. 174, WEISE, M. J. & INGRAM, V. M. (1976). Proteins and glycoproteins of membranes from developing chick red cells. J'. 610/. Chem. 251, ZENTGRAPH, H., DEUMLING, B., JARASCH, E. D. & FRANKE, W. W. (1971). J. biol. Chem. 246, (Received 21 April 1978)