J. Cell Sd. 69, 137-146 (1984)
137
Printed in Great Britain © The Company of Biologists Limited 1984
MICROFILAMENT ORGANIZATION AND TOTAL
ACTIN CONTENT ARE DECREASED IN HYBRIDS
DERIVED FROM THE FUSION OF HeLa CELLS WITH
HUMAN FIBROBLASTS
LINDA R. GOWING11, ROSS L. TELLAM2 AND
M. R. C. BANYARD
1
Department of Experimental Pathology and 2 Department of Physical Biochemistry,
The John Curtin School of Medical Research, The Australian National University,
P.O. Box 334, Canberra City, ACT2601, Australia
SUMMARY
The organization of the microfilaments and the actin content of matched pairs of tumorigenic and
non-tumorigenic HeLa/fibroblast hybrid cells was compared. Each pair consisted of a hybrid cell
line with suppressed tumorigenicity and a segregant tumorigenic cell line derived from it. The
tumorigenic HeLa parent cell line and a non-tumorigenic human fibroblast line were also included
in the comparison. Microfilament organization of the cell lines was assessed by fluorescence microscopy using NBD-phallacidin, a probe that specifically binds to actin filaments. The re-expression
of tumorigenicity is associated with a loss of microfilament organization. The actin content, as
measured by the DNase I inhibition assay, was significantly lower in the tumorigenic hybrids and
the HeLa parent than in the non-tumorigenic cells. The comparison was significant when the actin
concentration was expressed either per cell or per protein. Despite this reduced level of total actin
in tumorigenic cells, the ratio of monomeric to total actin remained constant in all cell lines tested.
INTRODUCTION
The cytoskeleton has been implicated in a wide range of cellular activities including
cell motility, secretion, pseudopod formation, contractile ring formation in mitosis
and the maintenance of cell shape and structure (Goldman et al. 1979; Brinkley,
1982; Weeds, 1982). The structural elements of the cytoskeleton consist of three
major components: namely, the microtubules, microfilaments and intermediate filaments. In view of the diversity of functions of the cytoskeleton it is not surprising that
there is increasing evidence for the involvement of the cytoskeleton, microfilaments
in particular, in transformation (Brinkley, 1982). Weber et al. (1974) have shown that
parallel arrays or bundles of microfilaments, called stress fibres, are reduced or absent
in transformed cells. Boschek et al. (1981) followed a time-dependent decrease in
microfilament organization during transformation of chick embryo fibroblasts with a
temperature-sensitive mutant of Rous sarcoma virus (RSV). Some of the strongest
evidence for the involvement of microfilaments in transformation comes from studies
of the RSV transforming protein, p60src. This protein kinase has been implicated in
the phosphorylation of tyrosine residues of the protein vinculin in transformed cell
lines (Sefton, Hunter, Ball & Singer, 1981; Hunter, 1980). Vinculin is located close
to membrane sites where microfilament bundles terminate and has been suggested as
138
L. R. Gowing, R. L. Tellam and M. R. C. Banyard
the link between actin bundles and the cell membrane (Weeds, 1982). In addition,
there has recently been evidence that vinculin is a possible substrate for protein kinase
C (Werth, Niedel & Pastan, 1983), which has in turn been implicated as the mediator
of the phorbol ester tumour promoters (Kikkawa et al. 1983). Further, Hendricks &
Weintraub (1981) have shown that the synthesis of a microfilament-associated
protein, tryopomyosin, is decreased in transformed cells.
The principal component of microfilaments is actin, a protein found ubiquitously
in animal cells. In non-muscle cells there is a dynamic equilibrium between actin
monomer (G-actin) and polymer (F-actin). This equilibrium probably facilitates
disassembly and re-assembly of the microfilaments at various positions and times
within a cell in response to the needs of the cell.
In view of the evidence for the involvement of microfilaments in transformation,
it is of interest to characterize both the structural organization of actin filaments and
the pools of actin monomer and polymer within cells that are similar except for the
expression of tumorigenicity. Human somatic cell hybrids (Klinger, 1980; Stanbridge et al. 1982) provide excellent material for such a comparison. These hybrid
cells have a morphology intermediate between those of the parents, allowing a clear
comparison to be made of the microfilament organization without the complexities
that arise from large variations in morphology.
MATERIALS
AND
METHODS
Cell culture
The HeLa/fibroblast hybrids and the D98 AH2 line were gifts from Stanbridge et al. and
Klinger. Characterization of these lines has been described previously (Klinger, 1980; Stanbridge
et al. 1982). The human fibroblast line (MRC-5) was obtained from Flow Laboratories and the
bronchiocarcinoma line (H.Ep.2) was obtained from the Sir William Dunn School of Pathology,
Oxford. All cell lines were routinely tested for tumorigenicity by the injection of 5 X 10* cells
subcutaneously into nude mice. Cell lines that produced tumours within 2-3 weeks of inoculation
were classified as tumorigenic. Cells were also regularly tested for contamination by mycoplasma
using the method of Chen (1977).
All cell lines were cultured in DMEM (Dulbecco's Modified Eagle's Medium) from Gibco. This
was supplemented with 5 % foetal calf serum, 5 % newborn calf serum (except in the case of the
MRC-5 line where 10% foetal calf serum was used), 2mM-pyruvate and antibiotics (50 units/ml
penicillin G, 50 units/ml streptomycin sulphate and 180 units/ml neomycin sulphate). Incubation
was at 37°C in a 5 % CO2 humidified atmosphere.
Microscopy
The microfilaments were fluorescently labelled with 7-nitrobenz-2-oxa-l,3-diazole-phallacidin
(NBD-phallacidin) essentially according to the method of Barak, Yocum, Nothnagel & Webb
(1980). Cells grown on 12 mm coverslips were washed free of medium with phosphate-buffered
saline (PBS) plus 0-02% sodium azide (Az). The cells were fixed with freshly prepared 4 %
paraformaldehyde in PBS/Az, extracted with acetone at — 20°C for 2-5 min and air-dried. Each
coverslip was covered with 30 jd of PBS/Az containing 5 ng NBD-phallacidin and left for 20 min
at room temperature. Controls were prepared by omitting the NBD-phallacidin. After rinsing twice
in PBS/Az and once in distilled water, the coverslips were mounted, cell-side down, in PBS/Az plus
20 % glycerol and sealed with clear nail varnish. Slides prepared in this way can be stored in the dark
at 4 °C for at least 5 months without any noticeable loss of fluorescence. Slides were examined under
oil immersion using a Leitz Orthoplan Universal Largefield microscope with a Ploemopak 2
Ac tin and tumorigenicity
139
Fluorescence vertical illuminator (filter block 12). For photography a Leitz Vario-Orthomat camera
system was used with Kodacolor VR1000 film.
Measurement of actin content
The total and monomeric actin content of cell suspensions was measured using the DNase I
inhibition assay of Blikstad et al. (1978) with the modifications of Blikstad & Carlsson (1982). This
assay is based on the principle that native monomeric actin specifically and stoichiometrically binds
to the enzyme DNase I and in so doing inhibits the latter's activity towards DNA. Actin polymer
does not significantly inhibit the activity of DNase I under the conditions used. The assay of DNase
I activity is based on the rate of increase of absorbance at 260 nm that results from the hydrolysis
of DNA by the enzyme. There is a linear relationship between the degree of inhibition of DNase
I activity and the actin monomer concentration for inhibition levels between 20 and 70%. An
appropriate standard curve in this range allows calculation of the actin monomer content of lysed
cell suspensions. The total actin content can be similarly determined after treatment of a sample of
the lysed cell suspension with O75 M-guanidine hydrochloride, which depolymerizes actin polymer
to native actin monomer.
The standard curve was prepared using monomeric actin from rabbit skeletal muscle prepared
according to the method of Pardee & Spudich (1983) with the Sephadex G-150 modification of
McLean-Fletcher & Pollard (1980). Polyacrylamide gel electrophoresis in the presence of sodium
dodecyl sulphate (SDS) indicated that this actin was greater than 98 % pure. Less than 3 % was nonpolymerizable as judged by the method of Tellam, Sculley, Nichol & Wills (1983). Standard curves
for the inhibition of DNase I activity versus actin monomer concentration were always performed
with freshly purified actin monomer. Lyophilization of actin monomer in the absence of stabilizing
agents such as sucrose results in substantial actin denaturation and therefore can result in artificially
low specific actin-inhibitory activity and high actin monomer contents of cell lysates (unpublished
observations).
The adherent cells were detached from the culture flasks by incubation in 0-025 % trypsin plus
0-02% EDTA in PBS for 10minat37 c C. Cells were pelleted by centrifugation (550#, 4min) and
resuspended in growth medium for a cell count. Cell counts were performed in duplicate using the
Trypan Blue exclusion method and a haemocytometer. The cells were pelleted, washed twice in
PBS and the final cell pellet was resuspended in the SF lysis buffer of Blikstad & Carlsson (1982).
This buffer (lOOmM-NaF; S0mM-KCl; 2mM-MgCl2; 1 miu-EGTA; lOrnM-potassium phosphate,
pH7-0; 0-2mM-dithioerythritol; 1 M-sucrose; 0'5% Triton X-100) has been reported to lyse cells
successfully and to maintain the balance of the actin monomer and polymer pools after lysis. We
found that the actin monomer content and the total actin content of a cell suspension treated in this
way was stable for about 30min and the results shown are means of three determinations made
during this 30-min period. The type of lysis buffer is crucial in the maintenance of stable pools of
actin monomer and polymer. Other lysis buffers (e.g. 10mM-Tris-HCl, pH8-0, 0-25 M-sucrose)
can monomerize the F-actin of a cell during lysis, while others can cause a time-dependent decrease
in the actin monomer content after cell lysis, presumably by actin polymerization (Blikstad &
Carlsson, 1982; author's observations). It should also be noted that cell lysate samples incubated
with guanidine hydrochloride for longer than 30min exhibited 'clumping' or aggregation making
measurements of DNase I activity difficult. Therefore, all measurements of DNase I inhibition were
made in the first 30min after addition of the guanidine hydrochloride to the sample.
The protein concentration in cell lysates was determined by the method of Lowry, Rosebrough,
Fair & Randall (1951), using bovine serum albumin as a standard and with the addition of 0-3 %
SDS to overcome the effects of Triton X-100. Actin concentrations for the DNase I assay standard
curve were measured spectrophotometrically using an extinction coefficient at 290 nm of 0-63 (1/g)
(Houk&Ue, 1974).
RESULTS
Assessment of actin organization byfluorescencemicroscopy
The microfilament organization of the cell lines was examined by fluorescence
microscopy and a summary of the results for all the cells examined is presented in
140
L. R. Gowing, R. L. Tellam and M. R, C. Banyard
Table 1. A qualitative summary of the features of the cell lines studied
Cell line
Status
Presence of
stress fibres
Presence of
intensities of
fluorescence
Source
MRC-5
H.Ep.2
Non-tumorigenic
Flow Laboratories
Tumorigenic
Sir William Dunn
School of Pathology
D98 AH2
Tumorigenic
Stanbridge
5E
Non-tumorigenic
Stanbridge
5L
Tumorigenic
Stanbridge
39E C13
Non-tumorigenic
Stanbridge
ESH39
Tumorigenic
Stanbridge
IA3CN2.1
Non-tumorigenic
Klinger
IA3 CN TG
Tumorigenic
Klinger
CN2B1 Col 1
Non-tumorigenic
Klinger
5A7mp 26.15
Tumorigenic
Klinger
Table 1. The analysis of these cell lines showed that tumorigenicity is associated with
the loss of microfilament organization (Fig. 1).
The MRC-5 cell line showed well-developed parallel arrays of stress fibres running
the entire length of the cell (Fig. 1A). These stress fibres were evenly distributed
throughout the cytoplasm and obscured the nucleus. This diploid fibroblast line is
similar to those used by Stanbridge & Klinger for the production of the somatic cell
hybrids that form the basis of this study.
The D98 AH2 cell line is a HeLa variant and was the tumorigenic parent used for
the production of the HeLa/fibroblast hybrids. The D98 AH2 cells are small and
rounded in shape and exhibit little microfilament organization (Fig. 1B). Only a small
proportion of the cells show stress fibres and when present these appear much finer
than those seen in the MRC-5 cells, and tend to be restricted to areas such as membrane ruffles and pseudopodia. Some D98 AH2 cells also contain 'intensities' of
fluorescence (see arrow in Fig. 1B) that are larger than adhesion plaques. The latter
are seen as small points of fluorescence when the focal plane is at the level of the
adherent cell surface.
All the non-tumorigenic hybrid cell lines showed well-organized stress fibres that,
although not as numerous as those seen in the MRC-5 cells, were more numerous and
more evenly distributed throughout the cells than those of the tumorigenic hybrids.
No intensities of fluorescence were seen in any of the non-tumorigenic cells.
~The tumorigenic cell lines were, within a narrow range, variable in morphology.
The cells tended to be more rounded and appeared smaller than the non-tumorigenic
cells. The intensities of fluorescence were found in large numbers in the 5L and ESH
39 lines but were present less frequently in the 5A7 mp 26.15 and IA3 CN TG lines.
Actin and tutnorigenicity
Fig. 1. A selection of photomicrographs representative of the cell lines studied. The
microfilaments are visualized with the fluorescent label NBD-phallacidin. A, MRC-5,
X200; B, D98 AH2, X200. The arrow indicates some intensities of fluorescence; c, 5E,
X160; D, 5L, X200; E, 39E C13, X200; F, ESH 39, X200; A,C,E, non-tumorigenic; B,D,F,
tumorigenic.
141
142
L. R. Govjing, R. L. Tellam and M. R. C. Banyard
Table 2. Summary of results and statistical analysis of DNase I inhibition assay of
actin content of cell lysates
Cell line
Status
% Monomeric
actin
% Total actin/unit Total actin/cell
protein
(Pg)
D98 AH2
32-8
3-2
8-98
5L
ESH39
32-4
2-7
9-68
>• Tumorigenic
36
3-3
IA3 CN TG
39-5
3-97
9-77
5A7mp 26.15,
36-7
3-15
8-16
MRC-5
40-3
4-4
17-3
5E
34-1
4-4
1601
37-5
4-7
33-1
5-6
14-97
15-1
35-2
5-2
11-5
-0-304
-5192
-5-115
39EC13
IA3CN2.1
* Non-tumorigenic
CN2B1 Col \J
/ statistic
Degrees of freedom
8
No significant
difference
8
Significantly
different with
P<0-001
10-8
8
Significantly
different with
P<0-001
The 5A7 mp 26.15 line is notable for the virtual absence of microfilament organization. The cells of the IA3 CN TG line were particularly variable in morphology and
microfilament organization.
Analysis of actin content by the DNase I inhibition assay
The actin content of the cell lines is summarized in Table 2, which also shows the
results of a <-test applied to the data grouped into tumorigenic and non-tumorigenic
phenotypes. The total actin content is significantly lower in the tumorigenic cell lines.
This is true whether the data are expressed as actin per cell or actin per protein. The
mean total actin content of tumorigenic cells is 65 % of the non-tumorigenic value.
This difference is not related to differences of cell density at confluence since actin
content was found to be independent of cell density (unpublished observations).
Despite the difference in total actin content the tumorigenic and non-tumorigenic cell
lines contained a remarkably constant ratio of monomeric to total actin (Table 2).
DISCUSSION
Our studies of these HeLa/fibroblast somatic cell hybrids have shown that the
tumorigenic cell lines can be distinguished from non-tumorigenic lines on the basis
of microfilament organization and actin content. All non-tumorigenic cells have highly
Actin and tumorigenicity
143
organized microfilaments with large numbers of stress fibres. In comparison, the
tumorigenic cells have a lower total actin content, fewer stress fibres and some cells
contain intensities of fluorescence. Since NBD-phallacidin specifically associates with
F-actin (Barak & Yocum, 1981) it is concluded that these observed intensities contain
F-actin.
The disorganization of microfilaments in transformed and tumorigenic cell lines
has been previously reported (Weber et al. 1974; Wang & Goldberg, 1976; Marshall,
Humphryes & Pollack, 1978; Der, Ash & Stanbridge, 1981; Boschek et al. 1981;
Stanbridge et al. 1982). The cause of this disorganization remains unclear but it may
be a function of adhesion and/or cell shape (Willinghamef a/. 1977; Der etal. 1981).
Indeed, when fibronectin or agents that raise intracellular levels of cyclic AMP are
added to transformed cells, the cells become more adherent and assume a morphology
resembling that of normal fibroblasts. This alteration in morphology is accompanied
by the reorganization of microfilaments (Puck, 1977; Deretal. 1981; Pastan, Willingham, de Crombrugghe & Gottesman, 1982). However, it is impossible to measure the
tumorigenicity of such 'reverse-transformed' cells since the change is not permanent
and is rapidly lost after removal of the agent causing the reverse transformation.
Watt, Harris, Weber & Osborn (1978), on the basis of a comparison of tumorigenic
and non-tumorigenic mouse hybrid pairs, concluded that microfilament disorganization was not a constant feature of tumorigenic cells. However, some of the hybrid cell
pairs, those derived from the PG19/diploid mouse fibroblast fusion, did show an
association of disorganized microfilaments with tumorigenicity. The cell lines
examined in our study were derived from a single tumour cell line (HeLa-D98 AH2)
and two different fibroblast cell lines. The consistent loss of microfilament organization of the tumorigenic hybrids in our study could reflect the nature of the fibroblasts
used to produce them or be a characteristic of human epithelial tumour hybrids. The
more general association of disorganized microfilaments with tumorigenicity in
human epithelial hybrid cells requires further study.
It seems that there are two identifiable components of the loss of microfilament
organization, one attributable to morphology and the other to tumorigenicity.
Because the morphology of the HeLa/fibroblast hybrids is intermediate between
those of the parents, gross differences in the tumorigenic and non-tumorigenic cells
due to morphology are eliminated and any remaining differences reflect alterations
due to the appearance of tumorigenicity. 'F-actin intensities' only appear in
tumorigenic cells and seem to reflect the inability of the tumorigenic cells to maintain
highly organized stresss fibres. Carley, Barak & Webb (1981) have previously reported the presence of F-actin aggregates in a variety of transformed cells after using
NBD-phallacidin to fluorescently label microfilaments. These aggregates or intensities may be caused by the low level of actin within the tumorigenic cells or by an
abnormality in actin-binding proteins resulting in a failure to stabilize the higherorder organization of the microfilaments.
The DNase I inhibition assay results showed that tumorigenic cells have reduced
levels of total actin. The actin content of chronic lymphocytic leukaemic lymphocytes
has been found to be lower than that of normal lymphocytes, i.e. l-4±0-3mg
144
L.R. Goiving, R. L. Tellam and M. R. C. Banyard
9
actin/10 cells and 4-3 % ± 1-1 % of the total cellular protein in leukaemic lymphocytes compared to 2-2 ± 0-4mg/109 cells and 6-6 % ± 1-8% of the total protein in
normal lymphocytes (Stark et al. 1982). The magnitude of reduction in leukaemic
lymphocytes (35 %) is similar to that reported here. Varani, Wass&Rao (1983), using
transformed lines of lymphoid and myeloid origin in a comparison with normal lymphocytes and leucocytes, also found the transformed cells to have a lower total actin
content (as a percentage of the total protein). The reduction in actin content of
tumorigenic cells is unlikely, on its own, to have a marked effect on the degree of actin
polymerization, since the actin concentrations within the cell should still be above the
critical actin concentration required for polymerization (Korn, 1982). Since the
reduction in total actin content could occur either by a balanced loss of actin isomers
or by a selective loss of particular isoforms, it is worth considering the mechanism by
which the actin content is regulated. Witt, Brown & Gordon (1983) showed that chick
embryo fibroblasts transformed by RSV had reduced amounts of a-actin while levels
of other major isoforms (^3 and y) remained constant. The mechanism of loss of actin
in the tumorigenic human somatic cells is unknown and is currently under investigation.
Since the ratio of monomeric to total actin was constant in all the cell lines while
total actin content decreased, the tumorigenic cells must contain reduced levels of
F-actin. The DNase I inhibition assay can only be used to measure G-actin and total
cell actin, with F-actin being determined by the difference between these values. The
assay does not therefore distinguish between orders of organization of F-actin. The
loss of microfilament organization with tumorigenicity may be a result of the
decreased amount of F-actin available for the formation of the higher-order structures
or it may be the result of alterations in the various actin-binding proteins that regulate
the formation of such structures. (For a recent review of these actin-binding proteins
see Weeds (1982).) The appearance of F-actin intensities with tumorigenicity
strengthens the idea of abnormal regulation of F-actin organization since these intensities must be some form of aberrant F-actin assembly. Two actin regulatory proteins,
tryopomyosin and vinculin, have been found to be altered in transformation:
tryopomyosin levels are decreased (Hendricks & Weintraub, 1981) while vinculin is
phosphorylated (Sefton et al. 1981; Werth et al. 1983). The importance of these
proteins and others in the expression of tumorigenicity is being investigated.
We thank Miss R. Stokes and Miss J. A. Turner for their excellent technical assistance and Mrs
K. Rabl for thefinaltyping of the manuscript.
REFERENCES
BARAK, L. S. & YOCUM, R. R. (1981).
7-Nitrobenz-2-oxa-l,3-diazole(NBD)-phallacidin:
synthesis of a fluorescent actin probe. Analyt. Biochem. 110, 31-38.
BAJIAK, L. S., YOCUM, R. R., NOTHNAGEL, E. A. & WEBB, W. W. (1980). Fluorescence staining
of the actin cytoskeleton in living cells with 7-nitrobenz-2-oxa-1,3-diazole-phallacidin. Proc. natn.
Acad. Set. U.SA. 77, 980-984.
BLIKSTAD, I. & CARLSSON, L. (1982). On the dynamics of the microfilament system in HeLa cells.
J. Cell Biol. 93, 122-128.
Actin and tumorigenicity
145
BLIKSTAD, I., MARKEY, F., CARLSSON, L., PERSSON, T . & LINDBERG, U. (1978). Selective assay
of monomeric and filamentous actin in cell extracts, using inhibition of deoxyribonuclease I. Cell
15, 935-943.
BOSCHEK, C. B., JOCKUSCH, B. M., FRIIS, R. R., BACK, R., GRUNDMANN, E. & BAUER, H.
(1981). Early changes in the distribution and organization of microfilament proteins during cell
transformation. Cell 24, 175-184.
BRINKLEY, B. R. (1982). The cytoskeleton: an intermediate in the expression of the transformed
phenotype in malignant cells. In Chemical Carcinogenesis (ed. C. Nicolini), pp. 435-467. New
York: Plenum Press.
CARLEY, W. W., BARAK, L. S. & WEBB, W. W. (1981). F-actin aggregates in transformed cells. J.
Cell Biol. 90, 797-802.
CHEN, T. R. (1977). In situ detection of mycoplasma contamination in cell cultures by fluorescent
Hoechst 33258 stain. Expl Cell Res. 104, 255-262.
DER, C. J., ASH, J. F. & STANBRIDGE, E. J. (1981). Cytoskeletal and transmembrane interactions
in the expression of tumourigenicity in human cell hybrids. J. Cell Sci. 52, 151-166.
GOLDMAN, R. D., MILSTED, A., SCHLOSS, J. A., STARGER, J. & YERNA, M.-J. (1979). Cytoplasmic
fibers in mammalian cells: cytoskeletal and contractile elements. A Rev. Physiol. 41, 703-722.
HENDRICKS, M. & WEINTRAUB, H. (1981). Tropomyosin is decreased in transformed cells. Proc.
natn. head. Sci. U.SA. 78, 5633-5637.
HOUK, T. W. JR & UE, K. (1974). The measurement of actin concentration in solution: a comparison of methods. Analyl. Biochent. 62, 66-74.
HUNTER, T . (1980). Proteins phosphorylated by the RSV transforming function. Ce//22, 647-648.
KIKKAWA, U., TAKAI, Y., TANAKA, Y., MIYAKE, R. & NISHIZUKA, Y. (1983). Protein kinase C
as a possible receptor protein of tumor-promoting phorbol esters. J. biol. Chem. 258,
11442-11445.
KLINGER, H. P. (1980). Suppression of tumourigenicity in somatic cell hybrids. I. Suppression and
reexpression of tumourigenicity in diploid human X D98 AH2 hybrids and independent segregation of tumourigenicity from other cell phenotypes. Cytogenet. Cell Genet. 27, 254—266.
KORN, E. D. (1982). Actin polymerization and its regulation by proteins from nonmuscle cells.
Physiol. Rev. 62, 672-737.
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.
MARSHALL, C. ] . , HUMPHRYES, K. C. & POLLACK, R. E. (1978). Microfilament bundles, LETS
protein and growth control in somatic cell hybrids. .7. Cell Sci. 33, 191-204.
MCLEAN-FLETCHER, S. & POLLARD, T. D. (1980). Identification of a factor in conventional muscle
actin preparations which inhibits actin filament self-association. Biochent. biophys.Res. Contmun.
96, 18-27.
PARDEE, J. D. & SPUDICH, J. A. (1982). Purification of muscle actin. Meth. Enzym. 85, 164-181.
PASTAN, I., WILLINGHAM, M., DE CROMBRUGGHE, B. & GOTTESMAN, M. M. (1982). Aging and
cancer: cyclic 3'5'-adenosine monophosphate and altered gene activity. Natn. Cancer Inst.
Monogr. 60, 7-15.
PUCK, T. T. (1977). Cyclic AMP, the microtubule-microfilament system, and cancer. Proc. natn.
Acad. Sci. U.SA. 74, 4491-4495.
SEFTON, B. M., HUNTER, T., BALL, E. H. & SINGER, S. J. (1981). Vinculin: a cytoskeletal target
of the transforming protein of Rous sarcoma virus. Cell 24, 165-174.
STANBRIDGE, E. J., DER, C. T., DOERSEN, C.-J., NISHIMI, R. Y., PEEHL, D. M., WEISSMAN, B.
E. & WILKINSON, J. E. (1982). Human cell hybrids: analysis of transformation and
tumourigenicity. Science, N.Y. 215, 252-259.
STARK, R., LIEBES, L. F., NEVRLA, D., CONKLYN, M. & SILBER, R. (1982). Decreased actin
content of lymphocytes from patients with chronic lymphocytic leukemia. Blood 59, 536—541.
TELLAM, R. L., SCULLEY, M. J., NICHOL, L. W. & WILLS, P. R. (1983). The influence of
polyethylene glycol 6000 on the properties of skeletal-muscle actin. Biochent. J. 213, 651-659.
VARANI, J., WASS, J. A. & RAO, K. M. K. (1983). Actin changes in normal human and rat
leukocytes and in transformed human leukocytic cells. J. natn. Cancer Inst. 70, 805-809.
WANG, E. & GOLDBERG, A. R. (1976). Changes in microfilament organization and surface
topography upon transformation of chick embryo fibroblasts with Rous sarcoma virus. Proc. natn.
Acad. Sci. U.SA. 73, 4065-4069.
146
L.R. Gowing, R. L. Tellam and M. R. C. Banyard
WATT, F. M., HARRIS, H., WEBER, K. & OSBORN, M. (1978). The disruption of actin cables and
microtubules in hybrids between malignant and non-malignant cells, and in tumours derived
from them. J. Cell Sci. 32, 419-432.
WEBER, K., LAZARJDES, E., GOLDMAN, R. D., VOGEL, A. & POLLACK, R. (1974). Localization
and distribution of actin fibers in normal, transformed and revertant cells. Cold Spring Harbor
Symp. quant. Biol. 39, 363-369.
WEEDS, A. (1982). Actin-binding proteins - regulators of cell architecture and motility. Nature,
Lond. 296, 811-816.
WERTH, D. K., NIEDEL, J. E. & PASTAN, I. (1983). Vinculin, a cytoskeletal substrate of protein
kinase C. J. biol. Chem. 258, 11423-11426.
WILLINGHAM, M. C , YAMADA, K. M., YAMADA, S. S., POUYSSEGUR, J. & PASTAN, I. (1977).
Microfilament bundles and cell shape are related to adhesiveness to substratum and are dissociable
from growth control in cultured fibroblasts. Cell 10, 375-380.
WITT, D. P., BROWN, D. J. & GORDON, J. A. (1983). Transformation-sensitive isoactin in passaged
chick embryo fibroblasts transformed by Rous sarcoma virus. J. Cell Biol. 96, 1766—1771.
{Received 26 January 1984-Accepted 16 February 1984)