PROLIFERATION AND DIFFERENTIATION OF STEM CELLS 

 OF THE BLOOD-FORMING SYSTEM OF THE MOUSE 



James E. Till 



Department of Medical Biophysics, University of Toronto and 

 The Ontario Cancer Institute, Toronto 5, Ontario, Canada 



First, I'll describe the system with which 

 we've been working, and, then, in the remain- 

 ing time, I'll tell you something of what we've 

 been doing with it. The work I'll describe was 

 done in collaboration with Dr. Ernest A. 

 McCulloch and Dr. Louis Siminovitch. 



The method we use to detect and count 

 stem cells has been described in detail else- 

 where (1, 2). The method is based on the 

 transplantation of cells into heavily irradiated 

 recipient animals. The irradiation converts the 

 recipients into culture vessels in which the 

 transplanted cells can grow by destroying the 

 proliferative capacity of the animals' own cells. 

 Inbred strains are used to avoid transplantation 

 difficulties. One can regard this as a form of 

 cell culture in vivo. The irradiated recipient 

 is well designed for this purpose, since it has 

 a built-in temperature control, a built-in pH 

 control and a built-in medium supply. 



Cells are taken from a normal donor, 

 suspended, counted and injected intravenously 

 into the irradiated recipients. Cells from any 

 blood-forming tissue may be used; we usually 

 use marrow. If you wait 10 days and look at 

 the spleens of these animals, you find colonies 

 in their spleens. These are colonies of cells 

 that have grown up from those cells of the trans- 

 plant which lodged in the spleens of the irradi- 

 ated animals. When the colonies are fixed in 

 Bouin's solution, they turn yellow, and you can 

 count them very easily. The number you get is 

 proportional to the number of cells you put in. 

 We find about 10 colonies per spleen per 10 ^ 

 cells injected. Why is there this rather low 

 efficiency? We think it is because most of the 

 cells that go into the mouse are fully differen- 

 tiated, or almost fully differentiated; that is, 

 they are cells that don't have much prolifera- 

 tive capacity left. Certainly, they do not have 



enough to make a colony of this size which con- 

 tains something like a million cells. 



POLLARD: That's not straight dilution, is 

 it? 



TILL: No, we've measured that. About a 

 fifth go to the spleen, so it's not that only a 

 small number get there (3), The spleen is a 

 pretty efficient filter of cells put into the blood 

 stream. 



The other point I want to make is that the 

 relationship between the number of colonies 

 formed and the number of cells injected is 

 linear and extrapolates back through the origin. 

 This suggests that the colonies are formed by 

 single entities which lodge in the spleen (1, 2). 



POLLARD: Have you done the Poisson test 

 on this? 



TILL: Yes. The distribution of the number 

 of colonies per spleen, for colonies formed by 

 transplanted cells, appears to be a Poisson 

 distribution. 



If you look at the colonies using histological 

 methods, you find that they contain differentiated 

 cells. Thus, these colonies are not like the 

 colonies that are formed by bacteria, for ex- 

 ample, where the cell composition is fairly 

 uniform. The cell composition of these colonies 

 is heterogeneous; they contain differentiated 

 cells and often more than one kind of dif- 

 ferentiated cell. These differentiated cells are 

 blood cells or their precursors, that is, erythro- 

 blasts, granulocytes and megakaryocytes. So 

 this raises the question, are these colonies 

 formed by the differentiation of more than one 

 initiating cell or does this mixture arise from 

 a single precursor? It's a rather important 

 question, so we have tried to find out about this. 



Dr. Andrew Becker did these experiments 

 (4). What he did was to obtain spleen colonies 

 from irradiated marrow cells. The irradiation 



69 



