inactivates some of the colony-forming ability 

 of these cells, so he had to start with more 

 cells (about 100 times as many) in order to end 

 up with discrete colonies. Then individual col- 

 onies were picked out of the unfixed spleen. The 

 individual colonies were separately dispersed 

 and chromosome preparations were made and 

 examined. 



At the dose that was used (650 rads) about 

 10% of the colonies contained chromosome aber- 

 rations. The point of interest was that when- 

 ever a chromosomal aberration was found in a 

 dividing cell in a colony, the same aberration 

 was present in more than 95% of the other 

 dividing cells in that colony. The interpretation 

 was that the irradiation of a single precursor 

 had generated the chromosomal aberration which 

 was then passed on to all its descendants, and 

 that, in fact, the colony was a clone formed from 

 this original damaged precursor. With ionizing 

 radiation, aberrations are formed in a random 

 fashion, so each aberration that you get is dif- 

 ferent. He found a total of 8 marked colonies. 

 Each one had a different kind of aberration, but 

 all appeared to be clones because all the divid- 

 ing cells of each colony carried the marker 

 characteristic of that colony. 



This doesn't really settle the question of 

 whether the different kinds of differentiated 

 cells have come from a common precursor be- 

 cause here we have just looked at dividing cells. 

 On the other hand, most of the differentiated 

 cells no longer are dividing. We haven't con- 

 clusively proven that the different kinds of dif- 

 ferentiated cells we find in a single colony have 

 arisen from a single precursor but this is fairly 

 good indirect evidence. 



We've taken as a working hypothesis that 

 the spleen colonies are clones, and that these 

 clones are formed by some sort of precursor 

 cell which can differentiate in multiple ways. 

 What one would like to know is, what governs 

 that choice? What determines whether the cell 

 will give rise to erythrocytic precursors or 

 granulocytic precursors or megakaryocytes? 

 This is what we're primarily interested in. 



Perhaps I should just refresh your mem- 

 ories for a moment about the organization of a 

 renewal system like this. It's been postulated 

 for a long time that there exist stem cells which 

 have two functions: one is to maintain their own 

 numbers; the other is to begin the pathway of 

 differentiation so one gets a series of divisions 

 resulting in cells with different functional 

 characteristics, thus ending up with a wholly 

 differentiated, fully specialized cell such as the 



red blood cell. This latter cell can't divide; it 

 hasn't even got a nucleus. Its immediate pre- 

 cursors can divide a few times, but capacity for 

 proliferation is limited and they have begun to 

 differentiate and form hemoglobin. The ones 

 nearest the stem cell can divide a lot and contain 

 no hemoglobin. The fully differentiated cells are 

 continually lost from the system, and they must 

 be replaced somehow. Since they can't divide 

 themselves, they must be replaced by the divi- 

 sion of the precursor cells. The ultimate cell, 

 the cell that has no precursors after the em- 

 bryonic stages, is the stem cell. Although this 

 type of cell was postulated to exist, it proved 

 difficult to obtain clearcut ways of recognizing 

 it experimentally. The spleen-colony technique 

 appears to be one way of doing it. This method 

 makes use of the major function of the stem 

 cells, which is to proliferate, and demands that 

 they be able to multiply through a number of 

 cell generations sufficient to give rise to a 

 colony of cells that you can see with the naked 

 eye. It's an arbitrary criterion, so we probably 

 don't detect all the stem cells by this criterion; 

 but we do see some. It gives us a class of stem 

 cells that we can look at. 



If this is true, we would like to know what 

 regulates the proliferation and differentiation 

 patterns of the stem cells. The stem cell, if it 

 is really multipotent, has several choices open 

 to it. One can postulate four of these. First, it 

 may choose to proliferate so as to maintain its 

 own numbers. We could call this, for the lack of 

 a better term, "self-renewal," although there is 

 evidence that the daughter stem cells may not 

 be exactly like their parents (5). If the system 

 is to keep going, the stem cells need to be able 

 to perform some sort of self-renewal, because, 

 by definition, there is no precursor for them in 

 the adult animal. Second, the stem cell may 

 differentiate to give rise to cells of the red cell 

 series. Third, it may give rise to cells of the 

 granulocyte series. Fourth, it may give rise to 

 megakaryocytes. Thus, four different pathways 

 of proliferation or differentiation are available 

 to the stem cell. One would like to find out what 

 governs whether or not a particular choice is 

 made. 



How can one go about trying to solve this 

 problem? Professor Pollard suggested this 

 morning that one way of approaching a problem 

 of this type is by the use of genetic methods, and 

 this is what we have tried to do. If you can find 

 a single gene mutation that affects some step 

 in the process, then you can assume that the 

 molecular basis for that particular effect is 



70 



