Ill RELATIONSHIP OF GROWTH in vUro 563 



cytes, cells resembling fibroblasts, mesothelium and even epithelium. The appear- 

 ance of cells may in some instances be more a fianction of the medium which sur- 

 rounds them than of the origin of the cells themselves. Parker (1950) commenting 

 upon Carrel's observations, remarks that "blood serum impresses upon leucocytes 

 the physiological and pathological peculiarities of the individual providing the 

 serum". Weiss applies the term "modulation" to the "loss of some actual functions 

 without change of potencies" (Bloom, 1937; Weiss, 1939), in contrast to true 

 dedifferentiation, which is the "loss of some actual functions and an increase in 

 potential functions". It is in fact perhaps better to think of the variable in cellular 

 differentiation, not as the cell as a whole, but as the particular cellular function or 

 functions which can be demonstrated. As Harrison (1940) said: "That the isolated 

 cells often achieve so little is no indication of limitation in what they can do, since 

 the conditions under which they are isolated are too restricted. Until such cells 

 are tested under a great variety of conditions, or better still all such as are possible, 

 their full capacity will not be known". 



Persistence of function, and the retention of characteristics peculiar to the 

 explanted tissue, are common features in cultures, even after many generations 

 of rapid proliferation and even though the cells undergo drastic morphological 

 changes which make it at first sight improbable that the specific properties are 

 still inherent in the new generations of cells. For example, immunological 

 specificities are retained (Landsteiner and Parker, 1940); ability to produce 

 hormones persists (Anderson and Haymaker, 1935-36); and liver cells are still 

 able to store glycogen (Westfall, Evans, Shannon and Earle, 1953). The most 

 numerous and convincing examples of the persistence of specific characters are 

 the tumour cells. After many generations in tissue culture such cells may still 

 produce, on inoculation into susceptible hosts, tumours histologically indis- 

 tinguishable from the parent neoplasms. There exist exceptions to this generaliza- 

 tion. In the first place, in a number of cases, normal cells after prolonged cultivation 

 in vitro, have undergone alteration and acquired the capacity to produce tumours 

 (Earle, 1943; Gey, Gey, Firor and Self, 1949; Goldblatt and Cameron, 1953; 

 Landschiitz, 1953). Secondly, the ability of some cultures of tumour cells to grow 

 in appropriate host animals has been lost or greatly attenuated (Sanford, Likely 

 and Earle, 1954). However, the great majority of tumour cell cultures retain for 

 months or years the capacity to induce tumours like those from which the cul- 

 tivated cells originated. 



While most cell strains, grown in tissue culture over a period of years, have 

 remained stable in specific properties, e.g. in ability to produce specific substances 

 (hormones by parathyroid, glycogen by liver cells), or in ability to produce 

 tumours in homologous animals, others, kept continuously multiplying, have 

 altered in their growth behaviour and have taken on new characteristics (Gey 

 and Bang, 1949; Parker, 1955; Gey, 1956), e.g. ability to multiply rapidly in 

 fluid suspensions, or ability to produce tumours. Acquisition of a new character 

 {e.g. malignancy) does not exclude the possibility of loss by the same cells of some 

 other property which might be examined {e.g. ability to produce fibres). In sum- 

 mary it may be said that stability is more common than change of potencies in 

 vitro; that in most cases little or nothing is known about factors determining 



Literature p. 581 



