J. Cell Sci. 30,87-97(1978)
87
Printed in Great Britain © Company of Biologists Limited 197S
COMMON FEATURES OF COATED VESICLES
FROM DISSIMILAR TISSUES: COMPOSITION
AND STRUCTURE
JOHN W. WOODS, MICHAEL P. WOODWARD
AND THOMAS F. ROTH
Department of Biological Sciences, University of Maryland,
Baltimore County, Catonsville, Maryland 21228, U.S.A.
SUMMARY
Coated vesicles were purified and characterized from porcine brain and chicken oocyte.
Electrophoresis on sodium dodecylsulphate (SDS) gels showed that coated vesicles from either
source have three major proteins in common with apparent molecular weights of 180000,
120000, and 55000 Daltons. Negatively stained specimens from both sources appear to consist
of a highly ordered array of short interconnected rods or ridges of material on the exterior
surface of a membrane vesicle. Coated vesicles purified from porcine brain and chicken oocyte
have mean external diameters of 75-0 and 85 oran,respectively. In coated vesicles of the
appropriate orientation, portions of the coat material appear to be organized into hexagons
and pentagons. Based on the observed variability of size and apparent structure, it is postulated
that the basic structural subunits of coated vesicles are the short rods or ridges of material
observed on the exterior membrane vesicle surface. It is suggested that multiples of these subunits can be assembled into many arrangements to yield coated vesicles of different sizes and
coat structure. It is also proposed that the structural subunit is a complex of 3 proteins of molecular weight of 180000, 120000 and 55000 Daltons.
INTRODUCTION
Membrane vesicles with a highly ordered array of material coating their cytoplasmic surface have been observed within the cytoplasm of many eucaryotic cell
types. The role of these coated vesicles in most cell types is not known; however,
several roles for coated vesicles in various tissues have been proposed. Roth & Porter
(1964) proposed that coated vesicles of developing oocytes mediate the specific transport of protein across the plasmalemma. Their proposal was based on the observation that there was a large increase in the number of coated vesicles in developing
mosquito oocytes when the oocytes were rapidly sequestering protein. They inferred
that this uptake was specific because of earlier work which demonstrated the specific
uptake of protein by developing oocytes in other species (Telfer, i960; Glass, 1961).
This postulate is supported by more recent evidence which demonstrates the specificity of protein transport into developing oocytes of mosquito and many other species
(Wallace & Dumont, 1968; Roth, Cutting & Atlas, 1976). Large numbers of coated
vesicles have also been observed in the developing oocytes of these species, suggesting
that they may play a role in this specific uptake. Coated vesicles seem to play a similar
role in human fibroblasts (Anderson, Goldstein & Brown, 1976). A low-density lipo-
88
J. W. Woods, M. P. Woodward and T. F. Roth
protein (LDL) cholesterol complex is bound to specific LDL receptors localized in
coated regions of the plasmalemma. These coated regions are believed to invaginate
and fuse to form coated vesicles which transport the LDL-cholesterol complex to
lysosomes for further processing.
Another role for coated vesicles has been postulated in presynaptic nerve terminals.
Studies of frog neuromuscular junction (Heuser & Reese, 1973) suggested that coated
vesicles mediate membrane recycling following neurotransmitter release. Coated
vesicles have been observed in guinea-pig brain, suggesting they may play a similar
role in mammalian nervous tissue (Kanaseki & Kadota, 1969). However, it is likely
that other mechanisms are also involved in presynaptic membrane recycling (Teichberg,
Holtzman, Crain & Peterson, 1975).
Recently it has become possible to characterize coated vesicles biochemically as
well as morphologically. This characterization was made possible by Pearse (1975),
who developed a reliable method for the purification of coated vesicles from crude
tissue homogenates. The Pearse procedure is a modification of a partial purification
scheme developed by Kanaseki & Kadota (1969). We have modified the Pearse (1975)
procedure and applied it to chicken ovaries and porcine brain in order to examine
coated vesicles from these phylogenetically and physiologically dissimilar tissues.
MATERIALS AND METHODS
Materials
Chicken oocytes and porcine brains were obtained from a local slaughter house.
Acrylamide, bisacrylamide, and ammonium persulphate were obtained from Bio-Rad
Laboratories. Tris (hydroxymethyl) aminomethane (Tris), 2(2V-morpholino) ethane sulphonic
acid (MES), sodium dodecylsulphate (SDS), ethyleneglycol-6«-(/?-amino-ethyl ether)iV,iV'tetra-acetic acid (EGTA), dithiothreitol (DTT), iV,iV",iv'',.A/'''-tetramethyl-ethylenediarnine and
sucrose were purchased from Sigma Chemical Co. Other chemicals were reagent grade and
were purchased from J. T. Baker. Deionized water was used for all solutions. Ultracentrifugation was carried out in a Beckman L5-50 ultracentrifuge.
Methods
Purification of coated vesicles. Purification of coated vesicles was carried out by a modification
of Pearse's (197s) procedure at 0-4 °C in an isolation buffer of 01 M MES (pH 65) 1 mM
EGTA, 05 mM MgCl2, and 002 % sodium azide.
Porcine brains were either used fresh or stored at —70 °C. The white matter was discarded
and the grey matter cut into small pieces, suspended in 3 vol. of isolation buffer, homogenized
for 30 s at high speed in a Waring blender, and then centrifuged at 20000 g for 30 min. The
pellet was discarded and the supernatant centrifuged at 100 000 g for 1 h to yield a pellet of
crude coated vesicles.
Chicken oocytes, 1-3 cm in diameter, were slit and gently squeezed to remove most of the
yolk. The drained oocytes consist of the oocyte plasma membrane and cortical cytoplasm, an
overlying monolayer of follicular epithelial cells, and a thick layer of connective tissue which
contains blood sinuses. This material was suspended in 3 vol. of isolation buffer, homogenized
for 20 s at low speed in a Waring blender, and centrifuged at 5000 g for 5 min to pellet debris
consisting mainly of connective tissue. The pellet was washed once with isolation buffer to
increase the yield of coated vesicles. This wash was combined with the first supernatant fraction
and treated in the same manner as the brain homogenate to yield a pellet of crude coated vesicles.
To purify further coated vesicles from either porcine brain or chicken oocyte, the crude
coated vesicle pellet was suspended in a minimal volume of isolation buffer and layered on a
Composition and structure of coated vesicles
89
linear 20-60 % (w/w) sucrose gradient prepared with isolation buffer. After centrifugation in a
SW 27 rotor at 24000 rev/min for 60 h, i-s-ml fractions were pumped from the bottom of the
gradient. The pure coated vesicles, found at sucrose concentrations of 50-55 % (w/w), were
pooled, diluted with buffer, and centrifuged at 100000 g for 1-5 h. This pellet was resuspended
in a small volume of isolation buffer to yield a suspension of pure coated vesicles.
SDS-poIyacrylamide gel electrophoresis. SDS-polyacrylamide gel electrophoresis was carried
out according to the procedure of Maizel (1969) using 5 % acrylamide, 01 % SDS in 8 cm x
o-6 cm glass tubes. Samples of 10-150 mg protein were prepared by heating at 100 °C for
5 min in 1 % SDS, 10 % sucrose, 10 mM Tris-HCl (pH 80), 1 mM EDTA, 40 mM DTT, and
50 /tg/ml Pyronin Y.
Some samples were carboxymethylated by adding a 10-fold molar excess of iodoacetamide
to DTT following the initial heating. These samples were heated to 100 °C for an additional
5 min. Samples were electrophoresed at 6 mA per gel until the tracking dye was within 1 era
of the end of the gel, at which time the gels were fixed and stained overnight in 30 % isopropanol,
7 % acetic acid, and 0-03 % Coomassie blue. Gels were destained in 10 % acetic acid and scanned
at 560 nm on a Gilford spectrophotometer equipped with a linear transport device. Myosin,
/?-galactosidase, bovine serum albumin, bovine gamma globulin, aldolase, and lysozyme were
used as molecular weight standards.
Electron microscopy. A drop of material was placed on a 400-mesh, Formvar- carboncoated grid and allowed to air dry. Samples were negatively stained with 2 % uranyl acetate.
Micrographs were taken at an initial magnification of 20000 to 50000 times on an Hitachi HU-12
electron microscope.
RESULTS
Analysis of purity of coated vesicle preparations
More than 95 % of the particles larger than 30 nm in diameter in our coated vesicle
preparations have a highly ordered array of material on their external surface characteristic of coated vesicles (Fig. 1). Few membrane vesicles and no sheets of membrane were observed. We have also observed a variable number of uniformly electrontransparent objects from io-o to 50-0 nm in diameter in the background of our
negatively stained preparations (Fig. 1). We believe these objects are staining artifacts
because their number can be decreased by using freshly prepared grids and stain;
also they are always present on control grids containing no coated vesicle sample.
Structural comparison
Negatively stained coated vesicles from either porcine brain or chicken oocyte
appeared to consist of a highly ordered spherical array of short interconnected rods
surrounding a central core. This central core had a continuous range of electron
opacities which we group into 3 major catagories: those in which the central structures
are entirely electron-opaque, partially opaque, or entirely electron-lucent (Fig. 2). In
general, opaque central structures tended to obscure some of the order of the coat.
However, when the central structure is electron-transparent we are occasionally able
to observe the central portion of the coat structure which is normally obscured by the
core of electron-dense material.
Measurements of the external diameter of randomly selected populations of coated
vesicles purified from either source are shown as a histogram in Fig. 3. The mean
external diameters of coated vesicles purified from chicken oocyte and porcine brain
were 85-0 and 75-0 nm, respectively. Both populations of coated vesicles displayed
a range of external diameters from approximately 60 to 105-0 nm.
go
J. W. Woods, M. P. Woodward and T. F. Roth
The coat material of well imaged coated vesicles appears to be composed of an
array of short interconnected rods. The peripheral portions of the coat, in a few of the
images which have the appropriate orientation, appear to approximate radially
arranged hexagons or pentagons surrounding a central polygon (Fig. 4). These
appropriately oriented coated vesicles can be grouped into 4 general categories based
on the apparent structure of their coat material. The most numerous category has
a mean external diameter of 72-5 nm and appears to consist of 4 pentagons and 2
Fig. 1. Purified coated vesicles from chicken oocyte (A) and porcine brain (B).
Negatively stained with 2 % uranyl acetate, x 60000.
hexagons oriented around a central hexagon (Fig. 4A). Similar structures are observed
in large numbers among coated vesicle isolates from either guinea-pig brain or chicken
oocyte. A second category of coat structure has an external diameter of 6o-o nm and
appears to consist of 6 pentagons oriented around a central hexagon (Fig. 4 c). Coat
structures of this class are common in brain isolates and very rarely observed in oocyte
isolates. The coat structures of these 2 classes have 6-fold rotational symmetry as
shown by Markham rotation (Fig. 4B, D).
The third class of coat structures has an external diameter of 82-5 nm and appears
to consist of 8 small peripheral polygons surrounding a larger central polygon (Fig.
4 E) . The fourth class of coat structure has an external diameter of 90- o nm and appears
to consist of 10 small peripheral polygons surrounding a large central polygon (Fig.
Composition and structure of coated vesicles
91
4G). The latter 2 classes have 8- and 10-fold rotational symmetry (Fig. 4F, H) and
have thus far been identified only in isolates from chicken oocyte. The shape of the
small peripheral polygons is not clear from our micrographs but their size suggests
that they are similar to the hexagons and pentagons which are clearly observed in
smaller coated vesicles. The structure of the large central polygon is obscured in our
micrographs.
We have also observed that the size of the hexagons and pentagons appeared to be
relatively constant (Fig. 4). However, this is difficult to quantitate due to the flattening
of the spherical structure that occurs during negative staining.
Fig. 2. Negatively stained coated vesicles from chicken oocyte (A) and porcine brain
(B). Within any population, some of the isolated coated vesicles have core structure
which are electron-opaque, partially electron-opaque or electron-lucent, x 150000.
Protein comparisons
Proteins from porcine brain and chicken oocyte coated vesicles were compared by
SDS-polyacrylamide gel electrophoresis. Ten separate preparations of coated vesicles
from each source were examined. Solubilized coated vesicles from each source have
3 major proteins of similar mobilities (Fig. 5). These proteins have apparent molecular
weights of 180000, 120 ooo, and 55000 Daltons and comprise 40-60% of the total
protein of oocyte coated vesicles and 60-80% of the total protein of brain coated
vesicles. The difference in relative percentage between oocyte and brain coated
92
J. W. Woods, M. P. Woodward and T. F. Roth
vesicles is due in part to the presence of a fourth major protein of molecular weight
140000 Daltons in chicken oocyte coated vesicles. The 180000-Dalton protein was
always predominant and constituted from 20-30 and 40-50% of the total protein of
oocyte and brain coated vesicles, respectively. The 140000-Dalton protein made up
7-10% of the total protein of chicken oocyte coated vesicles. A protein of similar
mobility was not observed in porcine brain coated vesicles.
40-
30-
20 H
6
z
10 -
59
67
1
75
82
89
External diameter, nm
97
104
Fig. 3. Histogram of external diameter of coated vesicles from (shaded) porcine brain
and (unshaded) chicken oocyte. Measurements were taken on randomly selected micrographs with total enlargement of 135000.
Certain minor proteins isolated from both sources also showed similar mobilities.
These minor proteins had molecular weights of 97000, 84000, and 16000 Daltons and
each accounted for less than 5 % of the total stained material. However, since our gel
system does not accurately separate proteins of less than 30000 Daltons molecular
weight, the 16000-Dalton band we observe in both samples may not represent protein
of similar molecular weight. Many other minor proteins of dissimilar mobilities were
observed. These results are summarized in Table 1.
Carboxymethylation of sulphydryl groups following reduction by DTT had no
Fig. 4. Appropriately oriented negatively stained coated vesicles representative of
general categories of coat structure. A, the coat structure of a 72-5-11111 diameter vesicle
appears to be 4 pentagons and 2 hexagons (arrows) surrounding a central hexagon,
common in isolates from either porcine brain or chicken oocyte. c, coat structure of a
6o-o-nm diameter vesicle appears to be 6 pentagons oriented around a central hexagon,
identified only in isolates from porcine brain, E, coat structure of a 82'5-nm diameter
vesicle appears to be 8 small polygons surrounding a larger central polygon. G, coat
structure of a cjo-o-nm diameter vesicle appears to be 10 small polygons surrounding
a larger central polygon. E and G have been identified only in isolates from chicken
oocyte. B, D, F and H are Markham rotations showing rotational symmetry of A, c, E
and G, respectively, x 300000.
Composition and structure of coated vesicles
93
B
D
-\
H
CEL 30
94
J- W. Woods, M. P. Woodward and T. F. Roth
effect on these results. However, the apparent molecular weight of all bands was
dependent upon the amount of 180000-Dalton protein present. If the gels were overloaded with this protein, the mobilities of all bands including the 180000-Dalton band
was increased.
180
\
i
55
\
it
140 120
Fig. 5. Solubilized coated vesicle proteins from porcine brain (A) and chicken oocytes
(B). Proteins were electrophoresed on 5 % polyacrylamide gel in the presence of o-i %
SDS. Proteins were stained with Coomassie blue, and values are stated in kilo Daltons.
Table 1. Proteins observed by SDS-page of coated vesicles purified from
chicken oocyte and porcine brain
Oocyte coated vesicles
Mol. wt., kDaltons
% total protein
l§o
20-30
140
120
107
7-10
15-20
<5
97
84
55
<5
<5
5-10
50
<5
39
35
16
<5
<5
<5
Brain coated vesicles
Mol. wt., kDaltons
% total protein
180
43-52
J57
<S
151
<5
120
10-15
97
84
55
52
<5
<5
10-16
<5
47
42
<5
<5
16
<5
% total protein on gel determined from densitometric scan at 560 nm of Coomassie Blue
stained gels following electrophoresis. % total calculated as area under peak divided by total
area under profile multiplied by 100. Area determined by cutting out profile and weighing.
DISCUSSION
A comparison of the protein composition of coated vesicles purified from chicken
oocytes and pig brain reveals three major proteins in common. These proteins, with
molecular weights of 180000, 120000, and 55000 Daltons, comprise more than 50%
of the total protein of isolated coated vesicles. The diversity of our sources of coated
vesicles suggests that these proteins may be common to coated vesicles in general. The
Composition and structure of coated vesicles
95
relative constancy of the weight percentages of the major bands from preparation to
preparation and between species and tissue suggests that these 3 proteins may be
intimately involved in the structure of coated vesicles.
Similar observations of the protein composition of coated vesicles have been reported. Blitz, Fine &Toselli (1977) observed that proteins of molecular weight 180000,
100 000 and 55000 Daltons make up a large percentage of the protein of coated
vesicles purified from rabbit brain. Pearse (1975) also reported the presence of large
amounts of a 180000-Dalton protein in coated vesicles purified from porcine brain.
In addition there appears to be a significant amount of two other proteins on the gels
shown by Pearse (1975, plate II). These proteins appear to have molecular weights of
about 120000 and 55000 Daltons, though this is not explicitly stated.
Pearse (1976) has suggested that the 180000-Dalton protein, clathrin, is common to
all coated vesicles and is the sole coat protein. While we agree that the 180000-Dalton
band is the predominant protein, the potential importance of the other two should not
be overlooked. Results from our laboratory (Woodward & Roth, 1977) suggest that
the 180000-Dalton protein is one of the major structural elements of the coat. However,
the presence of at least two other proteins in large amounts suggests it is not the sole
structural element.
Oocyte coated vesicles have an additional major protein which does not appear in
the gel patterns of brain coated vesicles. This protein has an apparent molecular weight
of 140000 Daltons and represents 7-10 % of the total protein of purified oocyte coated
vesicles. The 140000-Dalton protein may be a membrane-bound receptor involved
in the specific protein transport believed to be mediated by oocyte coated vesicles.
Another possibility is that it is the lipovitellin subunit of the phosvitin-lipovitellin
protein complex, shown by Cutting & Roth (1973) to be transported in large amounts
into developing chicken oocytes. Similar amounts of the phosvitin subunit would not
be expected to be observed because phosvitin stains only transiently with Coomassie
Blue and is therefore difficult to observe by ordinary methods. However, we have no
direct evidence to support either possibility. In addition, some of the minor proteins
observed in our preparations of purified oocyte coated vesicles may represent subunits of other specifically transported proteins, for example IgG (Cutting & Roth, 1973).
Many minor proteins not shared by porcine brain and chicken oocyte coated
vesicles were also observed. Conceivably some of these proteins may be involved in
functions of coated vesicles specific to the different tissues. For example, some may
represent receptor proteins for specifically transported materials or the specifically
transported material itself. Other proteins may be membrane proteins unrelated to
coated vesicle function but specific for a given tissue. These membrane proteins may
copurify with coated vesicle proteins by virtue of their chance location in regions of
the membrane which invaginated and pinched off to form the coated vesicles. If
these minor bands do represent proteins which are involved in the specific functions
of different tissues, we would not expect them to be found in coated vesicles isolated
from these different tissues.
Our procedures yield a relatively homogeneous population of coated vesicles purified
from chicken oocytes and porcine brain. These coated vesicles range in size from
7-2
96
J. W. Woods, M. P. Woodward and T. F. Roth
approximately 6o-o nm to greater than ioo-o nm in external diameter. However, the
distribution of sizes is such that the mean external diameters of coated vesicles
isolated from chicken oocyte and porcine brain are 85-0 and 75-0 nm, respectively.
Rees, Bunge & Bunge (1976) have reported similar variations in coated vesicle diameter. Earlier workers had observed that there were 2 size classes of coated vesicles,
75-0 and 1 oo-o nm and suggested that they were localized in the Golgi body and plasma
membrane, respectively. However, this distinction appears to be somewhat artificial.
In general, oocyte coated vesicles are larger than brain coated vesicles, a difference
which may reflect their different functions.
In addition to the heterogeneity in size observed, there is variation in the morphology
of the central core of negatively stained coated vesicles. Our observations suggest
that coated vesicles with electron-transparent cores are spherical arrays of coat matter
which lack a central membrane vesicle, coated vesicles with electron-opaque cores
contain a central vesicle and those with a partially opaque core have a damaged or
partial vesicle. However, these suggestions are tentative and require further examination.
Our electron micrographs of negatively stained coated vesicles from porcine brain
and chicken oocyte appear very similar. In many cases the apparent structure of coated
vesicles from these two tissues was indistinguishable. In addition, our micrographs
are similar to those of earlier authors (Kanaseki & Kadota, 1969; Crowther, Finch &
Pearse, 1976). On the basis of these observations, as well as observations of thin
sections, several models for the structure of the highly ordered array of material
coating the cytoplasmic surface of coated vesicles have been presented (Kanaseki &
Kadota, 1969; Crowther et al. 1976; Ockleford, 1976). Every model thus far proposed
postulates that a spherical array of short interconnected subunits of equal length form
the array of material observed on the exterior surface of the coated vesicle. Depending
on the particular model, these subunits are arranged into a varying number of contiguous hexagons and pentagons. In order to account for the proposed arrangements
of hexagons and pentagons, we postulate that all coated vesicles are composed of
identical structural subunits. These subunits are able to associate with one another to
form spherical arrays which are composed of a variable number of hexagons and pentagons. This would give rise to coated vesicles of different coat structures. Also, given
that these structural subunits are all identical, coated vesicles of larger external
diameter must contain more subunits, as well as have a different coat structure from
that of coated vesicles of smaller external diameter. Our micrographs showing various
numbers of polygons surrounding a central axis of symmetry, as well as the observation of a continuous range of external diameters, provide direct evidence for this
postulate. Additional evidence is provided by our observation that all hexagons and
pentagons appear to be of uniform dimensions. Ockleford (1976) reports similar
results. In addition, this postulate can explain the variety of coat structures observed
by Crowther et al. (1976).
The available evidence is not yet sufficient to determine conclusively whether the
structural subunits are rods which form a sphere somewhat larger than the membrane
vesicle, as suggested by Kanaseki & Kadota (1969), or ridges of material along the
surface of the membrane vesicle, as suggested by Ockleford (1976). However, our
Composition and structure of coated vesicles
97
observation of what we believe are coat structures which do not contain an internal
membrane vesicle suggests that the basic structural subunits are rods of material which
are distinct from the vesicle itself.
The biochemical nature of the basic subunit of coated vesicles is not yet clear.
However, we expect these subunits to be associations of the 3 principal proteins which
thus far appear to be common to all coated vesicles.
We are grateful to Dr Carol D. Linden for critical reading of the MS and for many
useful suggestions. This investigation was supported in part by NIH Research Grant
No. HD09549 from the National Institute of Child Health and Human Development.
REFERENCES
R. G. W., GOLDSTEIN, J. L. & BROWN, M. S. (1976). Localization of low-density
lipoprotein receptor on plasma membrane of normal human fibroblasts and their absence in
cells from a familial hypocholesterolemia homozygote. Proc. natn. Acad. Sci. U.S.A. 73,
2434-2438.
BLITZ, A. L., FINE, R. E. & TOSELLI, P. A. (1977). Functional and compositional similarities
between brain coated vesicles and sarcoplasmic reticulum fragments. In Contractile Systems
in Non-Muscle Tissues (ed. S. V. Perry, A. Margreth & R. S. Adelstein), pp. 81-90. Amsterdam : North-Holland Publishing.
CROWTHER, R. A., FINCH, J. T. & PEARSE, B. M. F. (1976). On the structure of coated vesicles.
J. molec. Biol. 103, 785-798.
CUTTING, J. A. & ROTH, T. F. (1973). Changes in specific sequestration of protein during transport into the developing oocyte of the chicken. Biochim. biophys. Ada 298, 951-955.
GLASS, L. E. (1961). Localization of autologous and heterologous serum proteins in the mouse
ovary. Devi Biol. 3, 787-804.
HEUSER, J. E. & REESE, T. S. (1973). Evidence for recycling of synaptic vesicle membrane
during transmitter release at the frog neuromuscular junction. 3- Cell Biol. 57, 315-344.
KANASEKI, T. & KADOTA, K. (1969). Vesicle in a basket. 3- Cell Biol. 42, 202-220.
MAIZEL, J. V. (1969). Gel electrophoresis of proteins and nucleic acids. In Fundamental
Techniques in Virology (ed. K. Habel & N. P. Salzman), pp. 334-362. New York: Academic
Press.
OCKLEFORD, C. D. (1976). A three-dimensional reconstruction of the polygon pattern of
placental coated vesicle membranes. 3- Cell Sci. 21, 83-91.
PEARSE, B. M. F. (1975). Coated vesicles from pig brain: Purification and biochemical characterization. 3- molec. Biol. 97, 93-98.
PEARSE, B. M. F. (1976). Clathrin, a unique protein associated with intracellular transfer of
membrane by coated vesicles. Proc. natn. Acad. Sci. U.S.A. 73, 1254-1255.
ANDERSON,
REES, R. P., BUNGE, M. B. & BUNGE, R. P. (1976). Morphological changes in the neuritic
growth cone and target neuron during synaptic junction development in culture. 3- Cell
Biol. 68, 240-263.
ROTH, T. F., CUTTING, J. A. & ATLAS, S. B. (1976). Protein transport: A selective membrane
mechanism. ,7. supramolec. Struct. 4, 527-548.
ROTH, T. F. & PORTER, K. R. (1964). Yolk protein uptake in the oocyte of the mosquito, Aedes
aegypti L. J . Cell Biol. 20, 3i3~332TEICHBERG, S., HOLTZMAN, E., CRAIN, S. M. & PETERSON, E. R. (1975). Circulation and turnover of synaptic vesicle membrane in cultured fetal mammalian spinal cord neurons. 3- Cell
Biol. 67, 215-230.
TELFER, W. H. (i960). The selective accumulation of blood proteins by the oocytes of saturnid
moths. Biol. Bull. mar. biol. Lab., Woods Hole 118, 338-351.
WALLACE, R. A. & DUMONT, J. N. (1968). The induced synthesis and transport of yolk proteins
and their accumulation by the oocyte in Xenopus laevis. 3- cell. Physiol. 72, 73-84.
WOODWARD, M. P. & ROTH, T. F. (1977). Analysis and selective dissociation of coated vesicles
from porcine brain. Submitted to 3- Cell Sci.
(Received 23 June 1977)