518
A . A . M O S C O N A
studies due to the consistency of their cellular content and regular sizes.
Thus in aggregates compounded of embryonic-chick liver cells, and
embryonic-mouse and chick precartilage cells, the restituted hepatic
tissue consisted solely of chick-cells while the cartilage masses were of
chimaeric composition (Figs. 12 and 13). Connective tissue cells
were frequently found distributed in the stromal framework regardless
of tissue or species derivation. Whether this is due to their "heteroafnnity", low contact specificity, contact inhibition, or entrapment is
not known. In none of the cell combinations studied was there any
noticeable evidence of cell transformation; since cells could be histologically distinguished individually, by their species differences, it was
possible to conclude that, in spite of interspersion and proximity, they
preserved their original tissue identities and grouped accordingly. It
should be stressed that, in all these tests, developmentally "determined"
cells were used; the possibility is by no means excluded that with cells
from earlier embryos different results might be obtained.
Trinkaus (1961) studied the fate of aggregated embryonic chick-cells
by using two types of markers: retinal pigment cells with pigment
granules as a self-marker in conjunction with other cells labeled with
tritiated thymidine or S
3 5
. The advantages and limitations of isotopic
labeling in aggregation studies were examined and the need for more
than one marking method stressed. The pigment cells were heterotypically combined with cells from different embryonic-chick tissues and
co-aggregated by centrifugation. The pellets were organ-cultured for
one or more days. In all cases, a cell type-specific sorting out of cells was
found. The regional dispositions of the restituted cell groupings differed
according to combination: epidermal cells tended to converge superficially, cartilage cells in the center, and pigment cells usually in an
intermediate position. The significance of such "hierarchies" and their
possible relation to the mechanisms of sorting-out were discussed.
Selective grouping of embryonic cells takes place also when such
cells are co-aggregated with neoplastic cells; but the precision of sorting
out and of the topographies established vary—perhaps suggestively—in
different combinations (Dabrowska-Piaskowska, 1959; Moscona, 1957a,
1961a, 1962a). The areas of normal cells may become re-infiltrated by
neoplastic elements and the initial grouping pattern thus modified
(Paul, 1961). Aggregation by rotation lends itself well to precise
studies on aggregative properties of various tumor cells and on interactions of normal and neoplastic cells; it might prove useful in examining various aspects of cohesiveness and selective affinities in relation to
properties of tumor cells versus normal cells. The system should be also
adaptable to bioassay of factors that might affect, directly or indirectly,
adhesiveness and infiltrative properties of neoplastic elements.
A . A . M O S C O N A
studies due to the consistency of their cellular content and regular sizes.
Thus in aggregates compounded of embryonic-chick liver cells, and
embryonic-mouse and chick precartilage cells, the restituted hepatic
tissue consisted solely of chick-cells while the cartilage masses were of
chimaeric composition (Figs. 12 and 13). Connective tissue cells
were frequently found distributed in the stromal framework regardless
of tissue or species derivation. Whether this is due to their "heteroafnnity", low contact specificity, contact inhibition, or entrapment is
not known. In none of the cell combinations studied was there any
noticeable evidence of cell transformation; since cells could be histologically distinguished individually, by their species differences, it was
possible to conclude that, in spite of interspersion and proximity, they
preserved their original tissue identities and grouped accordingly. It
should be stressed that, in all these tests, developmentally "determined"
cells were used; the possibility is by no means excluded that with cells
from earlier embryos different results might be obtained.
Trinkaus (1961) studied the fate of aggregated embryonic chick-cells
by using two types of markers: retinal pigment cells with pigment
granules as a self-marker in conjunction with other cells labeled with
tritiated thymidine or S
3 5
. The advantages and limitations of isotopic
labeling in aggregation studies were examined and the need for more
than one marking method stressed. The pigment cells were heterotypically combined with cells from different embryonic-chick tissues and
co-aggregated by centrifugation. The pellets were organ-cultured for
one or more days. In all cases, a cell type-specific sorting out of cells was
found. The regional dispositions of the restituted cell groupings differed
according to combination: epidermal cells tended to converge superficially, cartilage cells in the center, and pigment cells usually in an
intermediate position. The significance of such "hierarchies" and their
possible relation to the mechanisms of sorting-out were discussed.
Selective grouping of embryonic cells takes place also when such
cells are co-aggregated with neoplastic cells; but the precision of sorting
out and of the topographies established vary—perhaps suggestively—in
different combinations (Dabrowska-Piaskowska, 1959; Moscona, 1957a,
1961a, 1962a). The areas of normal cells may become re-infiltrated by
neoplastic elements and the initial grouping pattern thus modified
(Paul, 1961). Aggregation by rotation lends itself well to precise
studies on aggregative properties of various tumor cells and on interactions of normal and neoplastic cells; it might prove useful in examining various aspects of cohesiveness and selective affinities in relation to
properties of tumor cells versus normal cells. The system should be also
adaptable to bioassay of factors that might affect, directly or indirectly,
adhesiveness and infiltrative properties of neoplastic elements.