5. T H E L O C O M O T O R Y B E H A V I O U R OF CELLS
189
movement of the whole cell ceases; (5) another ruffled membrane
appears or expands and draws the cell away in a new direction.
Theories of the mechanism of contact inhibition (Abercrombie,
1961a) assume that the ruffled membrane is either the sole locomotory
organ of the cell or at least dominates the locomotory mechanism; and
that there is a competitive or similar inhibitory relation between active
ruffled membranes on the same cell. It is then necessary to explain how
the leading ruffled membrane is inhibited by contact. O f the various
possibilities, the one least open to objection at present seems to be that
the adhesion between the cells itself inhibits the membrane. Curtis
(1960, 1962) has suggested a way in which this may happen.
Contact inhibition does not result from every encounter between
fibroblasts. If the density of cell population is high enough to provide
sufficient opportunities for contact, some overlaps of cell on cell will be
found. The number varies with conditions (Curtis, 1961) and with the
type of fibroblast concerned (Abercrombie, Lamont and Stephenson,
1964). Both cells in mitosis and degenerating cells seem to lose their
inhibitory capacity, but most of the failures of inhibition are on the part
of apparently normal interphase fibroblasts.
Contact inhibition occurs between fibroblasts of quite different
origin, e.g. between those from embryo chick heart and neonatal mouse
skeletal muscle (Abercrombie, Heaysman and Karthauser, 1957); but
quantitatively the extent to which fibroblasts of one kind overlap those of
another does not seem to be simply related to the frequency of homologous overlaps of the two kinds of fibroblast (Abercrombie et al., 1964)
which suggests some specificity in the reactions. The sarcoma cells that
have been tested against normal fibroblasts show no or much reduced
inhibition by the fibroblasts, and the same is probably true for inhibition
of the fibroblasts by the sarcoma cells (Abercrombie et al., 1957;
Abercrombie, 1961b). Though the problem has not been analysed
quantitatively, it is probable that sarcoma cells show diminished contact
inhibition towards homologous sarcoma cells; their piling up on each
other, which is an expected outcome of lack of mutual contact inhibition, has often been remarked (for instance by Temin and Rubin, 1958).
The behaviour of sarcoma cells with regard to contact inhibition has
implications for malignant behaviour in vivo (Abercrombie and
Ambrose, 1962).
Epithelial cells show a behaviour closely similar to the contact
inhibition of fibroblasts, but most of the analysis has been made in vivo,
epidermis or other epithelia grafted in place of epidermis being readily
accessible to observation. This work has been concerned with what
happens when two advancing sheets of epithelium meet; it will be
referred to below. It is enough to note here that a cell at the edge of the
189
movement of the whole cell ceases; (5) another ruffled membrane
appears or expands and draws the cell away in a new direction.
Theories of the mechanism of contact inhibition (Abercrombie,
1961a) assume that the ruffled membrane is either the sole locomotory
organ of the cell or at least dominates the locomotory mechanism; and
that there is a competitive or similar inhibitory relation between active
ruffled membranes on the same cell. It is then necessary to explain how
the leading ruffled membrane is inhibited by contact. O f the various
possibilities, the one least open to objection at present seems to be that
the adhesion between the cells itself inhibits the membrane. Curtis
(1960, 1962) has suggested a way in which this may happen.
Contact inhibition does not result from every encounter between
fibroblasts. If the density of cell population is high enough to provide
sufficient opportunities for contact, some overlaps of cell on cell will be
found. The number varies with conditions (Curtis, 1961) and with the
type of fibroblast concerned (Abercrombie, Lamont and Stephenson,
1964). Both cells in mitosis and degenerating cells seem to lose their
inhibitory capacity, but most of the failures of inhibition are on the part
of apparently normal interphase fibroblasts.
Contact inhibition occurs between fibroblasts of quite different
origin, e.g. between those from embryo chick heart and neonatal mouse
skeletal muscle (Abercrombie, Heaysman and Karthauser, 1957); but
quantitatively the extent to which fibroblasts of one kind overlap those of
another does not seem to be simply related to the frequency of homologous overlaps of the two kinds of fibroblast (Abercrombie et al., 1964)
which suggests some specificity in the reactions. The sarcoma cells that
have been tested against normal fibroblasts show no or much reduced
inhibition by the fibroblasts, and the same is probably true for inhibition
of the fibroblasts by the sarcoma cells (Abercrombie et al., 1957;
Abercrombie, 1961b). Though the problem has not been analysed
quantitatively, it is probable that sarcoma cells show diminished contact
inhibition towards homologous sarcoma cells; their piling up on each
other, which is an expected outcome of lack of mutual contact inhibition, has often been remarked (for instance by Temin and Rubin, 1958).
The behaviour of sarcoma cells with regard to contact inhibition has
implications for malignant behaviour in vivo (Abercrombie and
Ambrose, 1962).
Epithelial cells show a behaviour closely similar to the contact
inhibition of fibroblasts, but most of the analysis has been made in vivo,
epidermis or other epithelia grafted in place of epidermis being readily
accessible to observation. This work has been concerned with what
happens when two advancing sheets of epithelium meet; it will be
referred to below. It is enough to note here that a cell at the edge of the