14. C E L L A G G R E G A T E S
513
is therefore displayed most clearly in the first 36 h of the aggregate, i.e.
prior to the onset of these secondary events. This applies, essentially to
all aggregates. Chimaeric aggregates have been maintained in culture
for up to a month without noticeable incompatibility between the cells.
The response of chimaeric aggregates to grafting into embryonic or
adult hosts has not yet been adequately investigated.
It would be rash to speculate or generalize at present beyond what
was said above about the behavior of cells in isotypic aggregates.
However, certain possibly fruitful lines for future study suggest themselves. Particularly interesting might be the outcome of recombinations
of cells obtained from various mutant and normal donors; as proposed
elsewhere (Moscona, 1962b), by suitable choice of such material much
could be learned on the relation of contact specificities, and interactions of cells to gene-controlled processes. The pioneering work of
Hadorn, Anders and Ursprung (1959) on intermingling and recombination of trypsin dissociated cells from imaginal discs of Drosophila
indicates the usefulness of such efforts. Such approaches are based on
the likely assumption that the dynamics of cell contact are an aspect of
cell differentiation and that the phenomena of cell contact are related
to gene-controlled biosynthetic sequences in cells. The availability of
various mammalian and avian mutant strains should make it possible
to explore the possible correlation between metabolic patterns of cells
and their contact behavior. There is considerable evidence for genetic
control of contact mechanisms in micro-organisms in which this possibility has been adequately examined. Adhesion of mating types in
bacteria, fungi, and protista, selective attachment of phages to bacterial
cell-wall substances, the "aggregateless" mutants in slime moulds
(Sussman and Sussman, 1956), are examples of situations which may
turn out to be quite relevant to the problem of cohesion of tissue cells.
It is certainly of considerable interest that, in all these cases, attachment
is mediated by specific products at the cell surface, the properties of
which reflect the genetic make-up and functional state of the cells. The
view is heuristically attractive that contact interactions of tissue cells
may involve similarly mediated processes.
As pointed out, most work on aggregation has been designed to
minimize those effects which might conceivably modify the expected
developmental expressions of the cells (Trinkaus, 1956; Grobstein,
1959). There is evidence that under other conditions the cells can display
novel characteristics. Aggregates of neural-retina cells of chick embryos
maintained in a state of overcrowding, in the presence of cytolyzed
retinal cells, tend to form lens-like bodies, instead of the typical
rosettes of sensory cells (Moscona, 1957b) (Figs. 14-16). It is of interest
s
513
is therefore displayed most clearly in the first 36 h of the aggregate, i.e.
prior to the onset of these secondary events. This applies, essentially to
all aggregates. Chimaeric aggregates have been maintained in culture
for up to a month without noticeable incompatibility between the cells.
The response of chimaeric aggregates to grafting into embryonic or
adult hosts has not yet been adequately investigated.
It would be rash to speculate or generalize at present beyond what
was said above about the behavior of cells in isotypic aggregates.
However, certain possibly fruitful lines for future study suggest themselves. Particularly interesting might be the outcome of recombinations
of cells obtained from various mutant and normal donors; as proposed
elsewhere (Moscona, 1962b), by suitable choice of such material much
could be learned on the relation of contact specificities, and interactions of cells to gene-controlled processes. The pioneering work of
Hadorn, Anders and Ursprung (1959) on intermingling and recombination of trypsin dissociated cells from imaginal discs of Drosophila
indicates the usefulness of such efforts. Such approaches are based on
the likely assumption that the dynamics of cell contact are an aspect of
cell differentiation and that the phenomena of cell contact are related
to gene-controlled biosynthetic sequences in cells. The availability of
various mammalian and avian mutant strains should make it possible
to explore the possible correlation between metabolic patterns of cells
and their contact behavior. There is considerable evidence for genetic
control of contact mechanisms in micro-organisms in which this possibility has been adequately examined. Adhesion of mating types in
bacteria, fungi, and protista, selective attachment of phages to bacterial
cell-wall substances, the "aggregateless" mutants in slime moulds
(Sussman and Sussman, 1956), are examples of situations which may
turn out to be quite relevant to the problem of cohesion of tissue cells.
It is certainly of considerable interest that, in all these cases, attachment
is mediated by specific products at the cell surface, the properties of
which reflect the genetic make-up and functional state of the cells. The
view is heuristically attractive that contact interactions of tissue cells
may involve similarly mediated processes.
As pointed out, most work on aggregation has been designed to
minimize those effects which might conceivably modify the expected
developmental expressions of the cells (Trinkaus, 1956; Grobstein,
1959). There is evidence that under other conditions the cells can display
novel characteristics. Aggregates of neural-retina cells of chick embryos
maintained in a state of overcrowding, in the presence of cytolyzed
retinal cells, tend to form lens-like bodies, instead of the typical
rosettes of sensory cells (Moscona, 1957b) (Figs. 14-16). It is of interest
s