476
CLIFFORD GROBSTEIN
low stability, that large volumes cannot be properly altered. It would
be of considerable interest to know in more detail the difference between
the immediate microenvironments of cells in the Konigsberg experiment, and of cells under and not under cellophane in the experiment of
Rose et al. Clearly we are faced with relatively subtle differences which
can have great consequence for differentiative behavior.
The matter is particularly important since it is now abundantly clear
that usual conditions for long-term growth of dispersed cells in culture
can be not only anti-differentiative in the sense of reversing existing
differentiation, but can select against differentiated cell types (Zaroff,
Sato and Mills, 1961) and can allow—perhaps even promote—cellular
variation of other kinds which may be competitive with differentiation.
Observations justifying this statement have been reviewed recently in a
number of places (Moser, 1960; Salzman, 1961; Wallace, Orsi, Ritter
and Moyer, 1961; Grobstein, 1959a; Swim, 1959; Sato, Zaroff and
Mills, 1960) and need not be repeated here. It is important to note,
however, that in consequence the earlier widespread conclusion that
cells propagate in vitro true to differentiative type must be re-evaluated.
Conceptions about propagability and differentiation in culture have
passed through several phases. Initially, the morphological simplification of complex explants was interpreted as indicating dedifferentiation
to an embryonic state (Champy, 1913). This view was attacked and
demolished when it was demonstrated that, in some instances, apparently dedifferentiated cells could be induced to redifferentiate—and that
this always occurred in the original and never in alternative directions.
From this followed the assumption of propagability of differentiation in
some covert form via invisible and irreversible determinants. This
assumption has now been challenged on several grounds (Trinkaus,
1956; Grobstein, 1959a), not least of which is the methodological
inadequacy of earlier data alleged to establish the redifferentiation of
cells known to have propagated through a number of culture generations. More recent analyses, specifically addressed to the issue, do not
support the earlier widely accepted conclusion.
Holtzer, Abbott, Lash and Holtzer (1960) have examined the
differentiative behavior of chondrogenic cells of 10-day chick embryo
vertebrae. They emphasize that these represent a homogeneous population of chondrocytes, with a few chondroblasts. The cells were freed
from their matrix by enzymatic and mechanical treatment. The suspension in nutrient medium was either dispersed on a plasma clot
substrate or pelleted by centrifugation and organ-cultured on agar. In
the latter case chondrogenesis occurred as judged by formation of a new
matrix metachromatically stainable with thionin. The dispersed cells
grew with a calculated generation time of 34 h. At intervals they were
CLIFFORD GROBSTEIN
low stability, that large volumes cannot be properly altered. It would
be of considerable interest to know in more detail the difference between
the immediate microenvironments of cells in the Konigsberg experiment, and of cells under and not under cellophane in the experiment of
Rose et al. Clearly we are faced with relatively subtle differences which
can have great consequence for differentiative behavior.
The matter is particularly important since it is now abundantly clear
that usual conditions for long-term growth of dispersed cells in culture
can be not only anti-differentiative in the sense of reversing existing
differentiation, but can select against differentiated cell types (Zaroff,
Sato and Mills, 1961) and can allow—perhaps even promote—cellular
variation of other kinds which may be competitive with differentiation.
Observations justifying this statement have been reviewed recently in a
number of places (Moser, 1960; Salzman, 1961; Wallace, Orsi, Ritter
and Moyer, 1961; Grobstein, 1959a; Swim, 1959; Sato, Zaroff and
Mills, 1960) and need not be repeated here. It is important to note,
however, that in consequence the earlier widespread conclusion that
cells propagate in vitro true to differentiative type must be re-evaluated.
Conceptions about propagability and differentiation in culture have
passed through several phases. Initially, the morphological simplification of complex explants was interpreted as indicating dedifferentiation
to an embryonic state (Champy, 1913). This view was attacked and
demolished when it was demonstrated that, in some instances, apparently dedifferentiated cells could be induced to redifferentiate—and that
this always occurred in the original and never in alternative directions.
From this followed the assumption of propagability of differentiation in
some covert form via invisible and irreversible determinants. This
assumption has now been challenged on several grounds (Trinkaus,
1956; Grobstein, 1959a), not least of which is the methodological
inadequacy of earlier data alleged to establish the redifferentiation of
cells known to have propagated through a number of culture generations. More recent analyses, specifically addressed to the issue, do not
support the earlier widely accepted conclusion.
Holtzer, Abbott, Lash and Holtzer (1960) have examined the
differentiative behavior of chondrogenic cells of 10-day chick embryo
vertebrae. They emphasize that these represent a homogeneous population of chondrocytes, with a few chondroblasts. The cells were freed
from their matrix by enzymatic and mechanical treatment. The suspension in nutrient medium was either dispersed on a plasma clot
substrate or pelleted by centrifugation and organ-cultured on agar. In
the latter case chondrogenesis occurred as judged by formation of a new
matrix metachromatically stainable with thionin. The dispersed cells
grew with a calculated generation time of 34 h. At intervals they were