Plasma membrane vesicles are shed from monolayer cell cultures during incubation in low concentrations of formaldehyde and other sulphydryl blocking reagents. In both 3T3 and SV3T3 cells disulphide reducing agents, including dithiothreitol and mercaptoethanol, potentiate formaldehyde-induced vesiculation. Plasma membrane vesiculation is shown to be a temperature-dependent phenomenon which occurs optimally between 22 and 37 °C and at pH 7·0 to 7·5. Membrane shedding is an energy-dependent phenomenon, requiring monovalent and divalent cations and slightly hypertonic medium. Plasma membrane vesiculation is not affected by pretreatment of cells with inhibitors of protein synthesis, i.e. cycloheximide, nor by agents which disrupt the cytoskeleton.

Exposure of a variety of monolayer cell cultures to formaldehyde or other sulphydryl blocking reagents, including V-ethylmalemide and iodoacetate, has been shown to induce the shedding of plasma membrane vesicles (Scott et al. 1978). Morphological studies have shown that multiple vesicles are released from individual cells and that the vesicles are free of contamination with intracellular organelles. Vesicles were shown, however, to contain some cytosol components but these are removed by hypotonic lysis of the vesicles thus yielding a pure plasma membrane preparation. Biochemical data showed that vesicles were enriched 10-fold in the plasma membrane enzyme marker 5’nucleotidase. They also carry receptors for Concanavalin A and wheat germ agglutinin and show an enrichment in cholesterol and sphingomyelin. In addition, the vesicle-derived plasma membranes contained approximately 20 polypeptides with molecular weights from approximately 200 000 to 10000 Daltons.

In this paper we have studied the factors that affect the process of plasma membrane vesiculation. We report that plasma membrane vesiculation is a dose, time- and temperature-dependent phenomenon. The process requires divalent and monovalent cations and is potentiated by disulphide reducing agents and by hypertonic medium. Vesiculation is inhibited by reagents reported to deplete cellular ATP levels, by temperature reduction and by high and low pH.

All correspondence to be sent to: Robert E. Scott, M.D., at the above address.

Paraformaldehyde and Cytochalasin B were purchased from Aldrich Chemical Company, Milwaukee, WI. Formaldehyde was prepared from paraformaldehyde as previously described (Scott et al. 1978). Dithiothreitol, mercaptoethanol, D2O, toxyl-lysyl-choromethylketone (TLCK), colchicine, valinomycin and trypsin were obtained from the Sigma Corporation, St Louis, MO. A23187 was kindly provided by Dr Robert J. Hosley, Lilly Research Lab., Indianapolis, IN. All reagents were obtained in the highest grade available. Except where otherwise noted, all phosphate-buffered saline solutions contained 8 g NaCl, 0 · 2 g KC1, 1 · 15 g Na2HPO4.7H2O, 0 · 2 g KH2PO4, 0 ·11g CaCl2.2H2O and 0 ·10 g MgCl2.6H2O per litre at pH 7 · 4. This reagent is designated CMPBS.

Cell lines

3T3 and SV3T3 cells (a gift from Dr George Todaro) were grown in Dulbecco’s modified essential medium (DMEM) containing 10 % calf serum. No antibiotics were added. Cells were grown in humidified incubators containing 10 % CO2 in air as previously described (Scott, 1976; Scott et al. 1978).

Assay of factors influencing plasma membrane vesiculation

Experiments were performed on washed cell monolayers grown to between 60 and 80 % confluence in 60-mm Petri dishes (Falcon Plastics, ON, Canada). Untreated cells or cells exposed to various reagents were incubated in vesiculant solutions. Vesiculation was then scored on a scale of from —to + + + + by phase-contrast microscopy. In separate control studies, the quality of membrane protein released when vesiculation was scored from —to + + + + was determined. Vesicles shed from cells showing different degrees of vesiculation were decanted from 49O-cm2 roller bottles (Corning Glass Co., Corning, NY) and were sedimented by centrifugation at 30000 g for 30 min at 4°C. Vesicle pellets were washed three times in 50 mM Trisbuffered saline (pH 7·4) and once in CMPBS to remove adherent serum protein and formaldehyde. The quantity of protein released per 108 cells was determined by assay of protein both by the method of Lowry, Rosebrough, Farr & Randall (1951) and by the fluorescamine method (Bohlem, Stein, Dairman & Udenfriend, 1973; Udenfriend et al. 1976). Comparable results were obtained by both procedures. Bovine serum albumin was used as the standard for both protein assays. All results presented in this paper have been repeated with reproducible results in at least three separate experiments.

Quantitation of plasma membrane vesiculation

Plasma membrane vesiculation was empirically scored by phase-contrast microscopy on a scale of —to + + + +. The results show that + + + + vesiculation yielded approximately 500 μg vesicle protein/108 cells; that + + + vesiculation yielded ∼ 350 μg vesicle protein/108 cells and that + + vesiculation yielded approximately 200 μg vesicle protein/108 cells. Vesiculation scored at + + or less gave yields of less than 75 μg vesicle protein/108 cells. It was also determined in numerous analyses that 60% of the vesicle protein was membrane associated and 40% was soluble, i.e. cytosol. Since an excellent correlation was established between vesiculation scored by phase-contrast microscopy and by more tedious biochemical analysis which required excessive quantities of cells, we employed microscopic assays in these studies.

Effects of variation in experimental conditions on plasma membrane vesiculation

Optimum plasma membrane vesiculation was induced in 3T3 and SV3T3 cells by exposure to 25 mM formaldehyde-2 mM DTT prepared in CMPBS (400 mosmol) for 2 h at 37 °C. The process, however, is temperature dependent. Table I shows that both 3T3 and SV3T3 cells shed vesicles at 37 °C and that maximum vesiculation occurs more rapidly in 3T3 cells. Membrane shedding also occurs in 3T3 cells at 22 °C and at 18 °C whereas no significant vesiculation occurs in SV3T3 at these temperatures. Vesiculation was also affected by the medium in which vesiculants are prepared. Vesiculants prepared in CMPBS, DMEM or 10 mM HEPES buffer saline containing calcium ions (pH 7 · 4) gave optimum vesiculation. Significantly less vesiculation was, however, observed when the vesiculants were prepared in phosphate-buffered saline without calcium or magnesium, or in 10 mM Tris-buffered saline (pH 7·4). These data suggest that divalent cations may be required for optimum vesiculation. This latter possibility was confirmed by a series of studies which are summarized in Table 2. Calcium was found to be the cation most effective in promoting vesiculation. The data show that if 25 mM formaldehyde-2 mM DTT was prepared in phosphate-buffered saline alone or in HEPES-buffered saline (HBS), only slight vesiculation was observed. Mg2+, Mn2+ and Zn2+ failed to promote aldehyde-induced vesiculation significantly; however, Ca2+ (0 · 5 mM or greater) greatly potentiated vesiculation.

Table 1.

Time and temperature effects on vesiculation

Time and temperature effects on vesiculation
Time and temperature effects on vesiculation
Table 2.

Effects of cations on vesiculation

Effects of cations on vesiculation
Effects of cations on vesiculation

A variety of other experimental conditions were tested for their effect on vesiculation. The effect of medium osmolarity was tested, i.e. 150 and 1000 mosmol. Formaldehyde (25 mM)-DTT (2 mM) was prepared in the following solutions which contained Ca 2+ and Mg2+: (a) 10 mM HEPES buffer in which the osmolarity was adjusted with NaCl; (b) 10 mM HEPES buffer in which sucrose was used to adjust the osmolarity; and (c) 10 mM HEPES-buffered saline in which sucrose was employed to increase the osmolarity. The results showed that + + to + + + + vesiculation was observed in only saline-containing solutions with an osmolarity of 400 mosmol or greater. No vesiculation was observed in salt-free sucrose solutions. These observations support the view that monovalent cations are required for vesiculation in addition to calcium ions.

Next we determined the effects of variations in pH on vesiculation. The results showed that maximum vesiculation occurred at a pH between 7 to 7·5. Significantly less or no vesiculation was observed when the pH was above 8 or below 6.

We also performed experiments in which the interval of exposure of the cell monolayers to the standard formaldehyde-dithiothreitol vesiculant was varied from 1 min to 2 h. We found that 3T3 cells showed significant vesiculation following a 15–20 min exposure. Vesiculation in SV3T3 cells required incubation of cells with the vesiculant for between 30 and 60 min.

Differences in the response of 3T3 and SV3T3 cells to plasma membrane vesiculants

3T3 and SV3T3 cells, therefore, differ in their response to formaldehyde-DTT. 3T3 cells required a shorter exposure interval for promotion of significant vesiculation and a lower incubation temperature. Table 3 also shows that 3T3 cells differ from SV3T3 cells in their response to non-formaldehyde containing vesiculants. For example, JV-ethyl malemide induced significant vesiculation in 3T3 cells, but not in SV3T3 cells. This suggests that there may be differences in the relative affinity of such reagents for sulphydryl groups in 3T3 and SV3T3 cells or that there are differences in the availability of the sulphydryl groups in normal and transformed cells. Whatever the cause of this differential effect, it was markedly modified by inclusion of a disulphide reducing agent in the vesiculant solution. When 3T3 and SV3T3 cells were incubated with N-ethyl malemide, iodoacetate, cacodylate orp-chloromercuribenzoate plus 2 mM dithiothreitol, both cell types shed plasma membrane vesicles (Table 3).

Table 3.

Differential effect of potential membrane vesiculants on 3T3 and SV3T3 cells

Differential effect of potential membrane vesiculants on 3T3 and SV3T3 cells
Differential effect of potential membrane vesiculants on 3T3 and SV3T3 cells

Effect of disulphide-reducing agents on plasma membrane vesiculation

Disulphide reducing agents not only modify the response of cells to N-ethyl-malemide, cacodylate and PCMB, they also potentiate the ability of low doses of formaldehyde to induce vesiculation. The results show that 250 mM formaldehyde alone promotes maximum vesiculation in 3T3 cells but that lower doses were in-effective. In the presence of 2 mM DTT, however, significant vesiculation was Observed at formaldehyde concentrations as low as 12 · 5 mM. When the DTT concentration was raised to 10 to 20 mM significant vesiculation could be induced by 6 to 7 · 5 mM formaldehyde. DTT alone did not promote vesiculation at any doses tested. Other disulphide reducing agents, i.e. mercaptoethanol, were also found to potentiate vesiculation.

Effect of metabolic inhibitors, cytoskeleton disruptive agents and other factors on plasma membrane vesiculation

Dinitrophenol, sodium azide, potassium cyanide and sodium fluoride inhibit ATP generation by different mechanisms and the ability of these reagents to deplete cellular ATP levels in 3T3 and SV3T3 cells has been established (Vlodavasky, Inbar sc Sachs, 1973). We therefore tested the effects of these agents on plasma membrane vesiculation. The results show that all inhibitors significantly decreased vesiculation in both 3T3 and SV3T3 cells (Table 4).

Table 4.

Effect of metabolic inhibitors on vesiculation

Effect of metabolic inhibitors on vesiculation
Effect of metabolic inhibitors on vesiculation

We also tested whether plasma membrane vesiculation could be influenced by reagents reported either to stabilize (Cross & Spindel, 1960) or cause disorder (Read & McElhaney, 1976; Seeman & Roth, 1972; Trump, Croker & Mergner, 1971) in plasma membranes, i.e. D2O (5–50%) and low doses of ethanol (0 · 5 to 1 M), respectively. Since numerous changes in the cell surface can be induced by proteolytic enzymes we also pretreated cells with trypsin (0 ·01 %, 30 min, 37 °C) and with the protease inhibitors TLCK (5 to 500 μg/ml) prior to the addition of vesiculants. We found that exposure of cells to these reagents had no effect on vesiculation. Pretreating cells with cyclohexamide (1 and 50 μ g/ml) for 30 min also failed to effect vesiculation.

The possibility that vesiculation might be influenced by modulation of the interaction of the cytoskeleton with plasma membrane was also tested. 3T3 and SV3T3 were pretreated for 3 h at 37°C with Cytochalasin B (10 μg/ml) or with colchicine (10 − 6 M) —reagents which have been reported to disrupt microfilaments (Scott, Maercklein & Furcht, 1977) and microtubules (Furcht & Scott, 1975), respectively. Such treatments had no detectable effect on vesiculation.

The results show that formaldehyde is the most effective vesiculant and that its capacity to promote vesiculation is greatly potentiated by disulphide reducing agents, such as dithiothreitol, in both cell types. Membrane shedding in 3T3 cells occurs most rapidly at a temperature of 37 °C and is inhibited by reduced temperature, i.e. below 22 °C. Vesiculation in SV3T3 occurs more slowly and only at nearly physiological temperatures.

In both 3T3 and SV3T3 cells optimum vesiculation was achieved only in the presence of an adequate energy supply, at a pH between 7 · 0 and 7 · 5 and in slightly hypertonic medium. In addition, the data show that for both 3T3 and SV3T3 cells calcium and monovalent cations are required for aldehyde-induced membrane vesiculation. In related experiments we found that vesiculation could also be induced solely by exposure of 3T3 cells to a combination of a monovalent cation ionophore, valino-mycin and a Ca2+ ionophore, A23187 at pH 6 · 8 (Scott, unpublished observations).

The results also show that specific vesiculants have a differential effect on normal and transformed cells and that this effect is abolished when disulphide reducing agents such as dithiothreitol are included in the vesiculant solution. This observation suggests that differences in the content or organization of plasma membrane sulphydryl-containing polypeptides may exist in normal and transformed cells. Our results support the data presented by Hynes & Destree (1977) who showed that virally transformed cells do indeed contain decreased disulphide-bonded proteins.

It is not possible at the present time to integrate these observations to establish a definitive mechanism by which sulphydryl blocking agents promote the shedding of plasma membrane vesicles. The data presented in this paper, however, suggest that it may involve a complicated process associated with a change in membrane permeability which leads to the formation of a cell surface bleb. Thereafter in an energy- and calcium-dependent event the base of the bleb appears to contract in a manner independent of the cytoskeleton. In the presence of calcium ions and elevated temperature the membrane at the base of the bleb then fuses and a plasma membrane vesicle is shed.

This work was supported by the Mayo Foundation and a grant from the U.S. Public Health Service.

Bohlem
,
P.
,
Stein
,
S.
,
Dairman
,
W.
&
Udenfriend
,
S.
(
1973
).
Fluorometric assay of proteins in the nanogram range
.
Archs Biochem. Biophys
.
155
,
213
220
.
Cross
,
P. R.
&
Spindel
,
W.
(
1960
).
The inhibition of mitosis by deuterium
.
Ann. N. Y. Acad. Sci
.
84
,
745
754
.
Furcht
,
L. T.
&
Scott
,
R. E.
(
1975
).
Effects of vinblastine sulphate, colchicine and lumicol-chicine on membrane organization in normal and transformed cells
.
Expl Cell Res
.
96
,
271
282
.
Hynes
,
R. O.
&
Destree
,
A.
(
1977
).
Extensive disulfide bonding at the mammalian cell surface
.
Proc. natn. Acad. Sci. U. S. A
.
74
,
2855
2859
.
Lowry
,
O. H.
,
Rosebrough
,
N. J.
,
Farr
,
A. L.
&
Randall
,
R. J.
(
1951
).
Protein measurement with the Folin phenol reagent
.
J. biol. Chem
.
193
,
265
275
.
Read
,
B. D.
&
Mcelhaney
,
R. N.
(
1976
).
Influence of membrane lipid fluidity on glucose and uridine facilitated diffusion in human erythrocytes
.
Biochim. biophys. Acta
419
,
331
341
.
Scott
,
R. E.
(
1976
).
Plasma membrane vesiculation: A new procedure for plasma membrane isolation
.
Science, N. Y
.
194
,
743
745
.
Scott
,
R. E.
,
Maercklein
,
P. B.
&
Furcht
,
L. T.
(
1977
).
Plasma membrane intramembranous particle topography in 3T3 and SV3T3 cells: The effect of Cytochalasin B
.
J. Cell Sci
.
23
,
173
192
.
Scott
,
R. E.
,
Perkins
,
R. G.
,
Zschunke
,
M.
,
Hoerl
,
B.
&
Maercklein
,
P. B.
(
1978
).
Plasma membrane vesiculation in 3T3 and SV3T3 cells. I. Morphological and biochemical characterization
.
J. Cell Sci
.
35
,
229
243
.
Seeman
,
P.
&
Roth
,
S.
(
1976
).
General anesthetics expand cell membranes at surgical concentrations
.
Biochim. BLophys. Acta
225
,
2691
2698
.
Trump
,
B. F.
,
Croker
,
B. P.
&
Mergner
,
W. J.
(
1971
).
In Cell Membranes: Biological and Pathological Aspects
(ed. G. W. Richter &
D. G.
Scarpelli
), pp.
84
126
.
Baltimore
:
Williams and Wilkins
.
Udenfriend
,
S.
,
Stein
,
S.
,
Bohlen
,
P.
,
Dairman
,
W.
,
Leimgruber
,
W.
&
Weigele
,
M.
(
1976
).
Fluorescamine: A reagent for assay of amino acids, peptides, proteins and primary amines in the picomole range
.
Science, N. Y
.
178
;
871
872
.
Vlodavasky
,
I.
,
Inbar
,
M.
&
Sachs
,
L.
(
1973
).
Membrane changes and adenosine triphosphate content in normal and malignant transformed cells
.
Proc. natn. Acad. Sci. U. S. A
.
70
,
1780
1784
.