14. C E L L A G G R E G A T E S
517
structures, and nephric tissues were formed (Fig. 11). These findings
were confirmed for other cell combinations (Moscona, 1960, 1962c);
they suggested that, in the systems studied, the cells (a) retained
throughout dispersal and aggregation their original histotypic identities, and (b) following initial random clustering became arranged in
uniform groupings according to their functional characteristics. These
assumptions fell short of proof, however, because of the difficulty of
determining the identity of intermingled chick-cells which, though from
different tissues, looked much alike when dissociated or in early
aggregates. The possibilities of cell transformation or metaplasia
remained. The obvious requirement was to have cells marked so that
when commingled they would nevertheless be individually distinguishable.
T o be adequate for aggregation studies, a cell marker should (a) not
interfere with cell viability or behavior, (b) be sufficiently distinct for
detection in individual cells in small populations, (c) not transfer to
other cells and, if picked up incidentally, not mark them, (d) persist for
several days, i.e. not be rapidly modified or eliminated due to metabolic
turn-over and differentiation of the cells. Ideally, a stable self-marker
would meet these requirements. Pigment cells were successfully used in
cell-localization studies (Weiss and Andres, 1952; Trinkaus and Gross,
1961); however, only the pigment layer of the retina provides readily
and cleanly available pigmented cells; moreover, the possibility of
release of pigment granules from dissociated cells and their phagocytosis
by other cells is strong. Chromosomal markers, detectable at cell
division, present a sampling difficulty in small aggregates. Progress
towards identification of cells in aggregates was made by the use of
combinations of cells from different species (Moscona, 1957a) and of
cells labeled with tritiated thymidine (Trinkaus and Gross, 1961).
Persistent and distinct differences in size and staining properties of
mouse- and chick-cells and the mutual tolerance of these cells in culture
(Wolff, 1954; Wolff and Bresch, 1955) suggested the use of cells of one
species as markers in combinations with cells from the other. Aggregates
produced (by self-aggregation or by rotation) in various heterotypicinterspecific combinations (Moscona, 1957a^ 1960, 1961a, c ; Auerbach
and Grobstein, 1958) showed first randomly intermingled cells which
subsequently became organized into groups and arranged into tissues.
In contrast with the chimaeric fabrics produced in isotypic-interspecific
aggregates (see p. 511), in heterotypic-interspecific aggregates there
was a clear sorting-out according to histotype. The precision of discriminatory mechanisms responsible for selective grouping is strikingly
demonstrable in aggregates comprising cells from several tissues.
Rotation-compounded aggregates are particularly suitable for such
517
structures, and nephric tissues were formed (Fig. 11). These findings
were confirmed for other cell combinations (Moscona, 1960, 1962c);
they suggested that, in the systems studied, the cells (a) retained
throughout dispersal and aggregation their original histotypic identities, and (b) following initial random clustering became arranged in
uniform groupings according to their functional characteristics. These
assumptions fell short of proof, however, because of the difficulty of
determining the identity of intermingled chick-cells which, though from
different tissues, looked much alike when dissociated or in early
aggregates. The possibilities of cell transformation or metaplasia
remained. The obvious requirement was to have cells marked so that
when commingled they would nevertheless be individually distinguishable.
T o be adequate for aggregation studies, a cell marker should (a) not
interfere with cell viability or behavior, (b) be sufficiently distinct for
detection in individual cells in small populations, (c) not transfer to
other cells and, if picked up incidentally, not mark them, (d) persist for
several days, i.e. not be rapidly modified or eliminated due to metabolic
turn-over and differentiation of the cells. Ideally, a stable self-marker
would meet these requirements. Pigment cells were successfully used in
cell-localization studies (Weiss and Andres, 1952; Trinkaus and Gross,
1961); however, only the pigment layer of the retina provides readily
and cleanly available pigmented cells; moreover, the possibility of
release of pigment granules from dissociated cells and their phagocytosis
by other cells is strong. Chromosomal markers, detectable at cell
division, present a sampling difficulty in small aggregates. Progress
towards identification of cells in aggregates was made by the use of
combinations of cells from different species (Moscona, 1957a) and of
cells labeled with tritiated thymidine (Trinkaus and Gross, 1961).
Persistent and distinct differences in size and staining properties of
mouse- and chick-cells and the mutual tolerance of these cells in culture
(Wolff, 1954; Wolff and Bresch, 1955) suggested the use of cells of one
species as markers in combinations with cells from the other. Aggregates
produced (by self-aggregation or by rotation) in various heterotypicinterspecific combinations (Moscona, 1957a^ 1960, 1961a, c ; Auerbach
and Grobstein, 1958) showed first randomly intermingled cells which
subsequently became organized into groups and arranged into tissues.
In contrast with the chimaeric fabrics produced in isotypic-interspecific
aggregates (see p. 511), in heterotypic-interspecific aggregates there
was a clear sorting-out according to histotype. The precision of discriminatory mechanisms responsible for selective grouping is strikingly
demonstrable in aggregates comprising cells from several tissues.
Rotation-compounded aggregates are particularly suitable for such